From gergely.csatari at nokia.com  Fri Feb  1 15:57:49 2019
From: gergely.csatari at nokia.com (Csatari, Gergely (Nokia - HU/Budapest))
Date: Fri, 1 Feb 2019 15:57:49 +0000
Subject: [Edge-computing] Status of Keystone federation testing with
	tempest
In-Reply-To: <1547567769.2225.16.camel@suse.com>
References: <HE1PR07MB17223E0901521B324B751116E10F0@HE1PR07MB1722.eurprd07.prod.outlook.com>
 ,
 <HE1PR07MB17225D5FF9D1C67A5BEB4AFFE11B0@HE1PR07MB1722.eurprd07.prod.outlook.com>
 ,
 <HE1PR07MB1722055116AAF2366A575C5AE1190@HE1PR07MB1722.eurprd07.prod.outlook.com>
 ,
 <AM3PR07MB0743AFEEBFF769092A372E0AE1890@AM3PR07MB0743.eurprd07.prod.outlook.com>
 <AM3PR07MB0743CF9D2170D3AA93DCC7BDE1840@AM3PR07MB0743.eurprd07.prod.outlook.com>
 <1547567769.2225.16.camel@suse.com>
Message-ID: <AM6PR07MB61676F9D2AB9664175C2DBD5E1920@AM6PR07MB6167.eurprd07.prod.outlook.com>

Hi,

Thanks for the config manuals. The main difference between your setup and mine is that I’m using Shibboleth Idp while you are using Keystone as Idp (I’m doing K2S not K2K). The target in my case to be able to run this using DevStack.

In theory I could also try to use K2K, but I see only two options:

  1.  Use an other Keysone instance. For this we should be able tor un 2 Keystone instances in DevStack
  2.  Use the same Keystone instance. Is it possible, that the same Keysone instance federates with itself (this is a bit federation singluarity…)

And no, unfortunatelly I did not document anything I did. I planned to do it manually quick and automate the whole thing, so we do not need to document it. I plan to document my steps somewhere in the future.

Br,
Gerg0

From: Manuel Buil <mbuil at suse.com>
Sent: Tuesday, January 15, 2019 4:56 PM
To: Csatari, Gergely (Nokia - HU/Budapest) <gergely.csatari at nokia.com>; nick at stackhpc.com; knikolla at bu.edu; colleen at gazlene.net; edge-computing at lists.openstack.org; Ildiko Vancsa <ildiko at openstack.org>; Beierl, Mark <Mark.Beierl at dell.com>
Subject: Re: Status of Keystone federation testing with tempest

Hi Gerg0,

I had a K2K deployment working (I was able to fetch a token from the SP) and I tried to document all the steps:

https://wiki.opnfv.org/display/EC/Keystone+Federation+demo+in+opensuse

Regarding the IdP config, this is what I wrote to check that things were fine at the IdP side:

# If the IdP configuration is correct, you should see the following in /var/log/apache2/keystone_access.log when the user requests the token through the federated_project:

172.29.236.11 - - [07/Nov/2018:18:53:04 +0000] "GET /v3 HTTP/1.1" 200 254 "-" "openstacksdk/0.17.2 keystoneauth1/3.10.0 python-requests/2.19.1 CPython/2.7.13"

172.29.236.11 - - [07/Nov/2018:18:53:04 +0000] "POST /v3/auth/tokens HTTP/1.1" 201 6234 "-" "openstacksdk/0.17.2 keystoneauth1/3.10.0 python-requests/2.19.1 CPython/2.7.13"
172.29.236.11 - - [07/Nov/2018:18:53:04 +0000] "POST /v3/auth/OS-FEDERATION/saml2/ecp HTTP/1.1" 200 6535 "-" "openstacksdk/0.17.2 keystoneauth1/3.10.0 python-requests/2.19.1 CPython/2.7.13"

By looking at your logs it seems you are getting a config problem in Shibboleth, however, Keystone is capable of providing the SAML implementation for IdP, so in theory you don't need Shibboleth at the IdP side. Could you check the steps I followed in that link in the section "### MAKING EDGE1 IdP ###"? Have you documented the steps you have followed? Perhaps we could compare them.

Regards,
Manuel

On Thu, 2019-01-10 at 14:50 +0000, Csatari, Gergely (Nokia - HU/Budapest) wrote:

Hi,



Okay, the problem was with the wrong IP configuration for the metadata fetching.

Now it is sorted out, but Shibboleth is still not happy.



I've attached the error log I get now from the idp.



If you have any idea for the reason please notify me.



Br,

Gerg0




From: Csatari, Gergely (Nokia - HU/Budapest)
Sent: Monday, January 7, 2019 4:23:23 PM
To: nick at stackhpc.com<mailto:nick at stackhpc.com>; knikolla at bu.edu<mailto:knikolla at bu.edu>; colleen at gazlene.net<mailto:colleen at gazlene.net>; mbuil at suse.com<mailto:mbuil at suse.com>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; Ildiko Vancsa; Beierl, Mark
Subject: Re: Status of Keystone federation testing with tempest


Hi,



Just to send a signal, that I'm still working on this (with best effort).



What I could figure out in the last some months, is that we need a "verify=False" parameter in the session.post of send_identity_provider_authn_request (in saml2_client.py) to be able to connect to the IdP using self signed certificates.



Now I'm able to communicate with the IdP and I can see, that there is a problem with the IdP config.

This is the log of Shibboleth:
2019-01-07 11:33:55,165 - WARN [net.shibboleth.idp.profile.impl.SelectProfileConfiguration:111] - Profile Action SelectProfileConfiguration: Profile http://shibboleth.net/ns/profiles/saml2/sso/ecp is not available for RP configuration shibboleth.UnverifiedRelyingParty (RPID http://127.0.0.1/shibboleth)
2019-01-07 11:33:55,167 - WARN [org.opensaml.profile.action.impl.LogEvent:105] - A non-proceed event occurred while processing the request: InvalidProfileConfiguration
I keep debugging this and update you about the results.



Would it help if I would push all modified things to GitHub with a description how am I doing the configuration?





Br,

Gerg0

________________________________
From: Csatari, Gergely (Nokia - HU/Budapest)
Sent: Saturday, September 15, 2018 12:36 AM
To: nick at stackhpc.com<mailto:nick at stackhpc.com>; knikolla at bu.edu<mailto:knikolla at bu.edu>; colleen at gazlene.net<mailto:colleen at gazlene.net>; mbuil at suse.com<mailto:mbuil at suse.com>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: Re: Status of Keystone federation testing with tempest


Hi,



Some update on the open issues:

- Make this work 😉: Okay, I realized, that I use the wrong certificate in the Idp. Based on the IdP-s description I should generate a p12 certificate using the certificate and the key used by the Shibboleth Sp. When I try to generate the certificate I get a strange error:

openssl pkcs12 -inkey /etc/shibboleth/sp-key.pem -in /etc/shibboleth/sp-cert.pem -out ../keystone-shibboleth-idp-dockerized/shibboleth-idp/credentials/idp-browser.p12

139822780584384:error:0D0680A8:asn1 encoding routines:asn1_check_tlen:wrong tag:../crypto/asn1/tasn_dec.c:1129:
139822780584384:error:0D07803A:asn1 encoding routines:asn1_item_embed_d2i:nested asn1 error:../crypto/asn1/tasn_dec.c:289:Type=PKCS12

Google tells me that this is becouse one of my pem files are in the wrong format. This is really strange as the error persist even after I regenerated these files with

shib-keygen -f -y 1

   - Figure out how to start a Container in a Keystone plugin or a tempest plugin - no progress on this
   - Figure out ow to integrate with CI - no progress on this
   - Figure out how to use static certificates and keys, so the same IdP container image can be used.

If you are bigger fans of IRC than email I can start sending these updates to the keystone channel.

Br,
Gerg0


________________________________
From: Csatari, Gergely (Nokia - HU/Budapest)
Sent: Wednesday, September 12, 2018 11:25:32 PM
To: nick at stackhpc.com<mailto:nick at stackhpc.com>; knikolla at bu.edu<mailto:knikolla at bu.edu>; colleen at gazlene.net<mailto:colleen at gazlene.net>; mbuil at suse.com<mailto:mbuil at suse.com>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: Re: Status of Keystone federation testing with tempest


Hi,



Some update on the open issues.
   - Make this work 😉 - here I have some progress, however I can not explain why. Now keystone is able to reach Shibboleth and Shibboleth answers with FatalProfileException "A valid authentication statement was not found in the incoming message.". I continue to figure out what is the problem.
   - Set the idp address in the correct place - This is done thanks to gmann.
   - Figure out how to start a Container in a Keystone plugin or a tempest plugin - Here I try to use https://github.com/openstack/devstack-plugin-container however I'm not sure if this is the right tool to start containers in DevStack environment.
   - Figure out ow to integrate with CI - no progress on this

I'm still happy get any help either in mail, IRC or in person on the PTG.

Thanks,
Gerg0








________________________________
From: Csatari, Gergely (Nokia - HU/Budapest)
Sent: Friday, August 31, 2018 1:03:43 PM
To: nick at stackhpc.com<mailto:nick at stackhpc.com>; knikolla at bu.edu<mailto:knikolla at bu.edu>; colleen at gazlene.net<mailto:colleen at gazlene.net>; mbuil at suse.com<mailto:mbuil at suse.com>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: Status of Keystone federation testing with tempest


Hi,



I'm working on this for a while, but as I am not a big expert of IdP, Keysone or Tempest I have a bit slow progress. I decided to share what I did and what are my current probelms to 1) inform the team about the progress 2) keep a record for myself 3) hoping for help and/or hints.



So I did this:

1) Get an Ubuntu

2) Install devstack with
enable_plugin keystone git://git.openstack.org/openstack/keystone
enable_service keystone-saml2-federation
Here I already ran into some package maangement issues due to some libcurl3 and libcurl4 incompatibility issue what I solved using https://launchpad.net/~xapienz/+archive/ubuntu/curl34

3) Install the Keystone tempest plugin

4) Build a Shibboleth IdP container based on https://github.com/Unicon/shibboleth-idp-dockerized with the configuration I believe is correct. I have a feeling that we will need to set a proper organisation for this if we want to publish this to Docker Hub. By the way is there a container registry maintained in the OpenStack development infra?

5) Run the container and expose 8080, 4443 and 8443 ports

This is a half success. Shibboleth contacts Keystone (or actually the Shibboleth apache module) for metadata update, but it works only on the first attempt. The regular updates are not working for some reason.

Also I was not able to get a positive answer from the status script of Shibboleth itself, so i just decided to move a bit forward.

6) Set idp_url to https://localhost:8080/idp/profile/SAML2/SOAP/ECP in _request_unscoped_token inside the Keystone tempest plugin. Here I have no idea where the configuration is actually stores and where should I set this in a nice way.

7) Run the tempest tests. Now here I get an error message which tells me about SSL version numbers (hands.hake: Error([('SSL routines', 'ssl3_get_record', 'wrong version number')],)",),))). I tried to use different ssl versions with Curl, but it complains about the lack of support in libsso.



So here I am now.



Things what I deffinetly should figure out:

   - Make this work 😉

   - Set the idp address in the correct place

   - Figure out how to start a Container in a Keystone plugin or a tempest plugin

   - Figure out ow to integrate with CI


Any comments are welcome.

Br,
Gerg0

Curl 3 and 4 : Evgeny Brazgin - launchpad.net<https://launchpad.net/~xapienz/+archive/ubuntu/curl34>
launchpad.net
PPA contains libcurl4 package, which supports both libcurl3 and libcurl4 API.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190201/4e7b4fed/attachment-0001.html>

From ildiko at openstack.org  Thu Feb  7 21:07:00 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Thu, 7 Feb 2019 22:07:00 +0100
Subject: [Edge-computing] ETSI MEC Overview on the weekly call next week
Message-ID: <75CAFA50-5A0D-46C0-A940-550AA6D66F0A@openstack.org>

Hi,

A quick reminder that next week we will start the weekly call with an update about ETSI MEC next week and we can also explore ways to collaborate with the group.

The call will be in the US-friendly time slot at __7am PT / 1500 UTC__ on Tuesday, February 12.

You can find the meeting details here: https://wiki.openstack.org/wiki/Edge_Computing_Group#Meetings

Please let me know if you have any questions.

Thanks and Best Regards,
Ildikó




From ildiko at openstack.org  Tue Feb 12 14:06:45 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Tue, 12 Feb 2019 15:06:45 +0100
Subject: [Edge-computing] Meeting in less than an hour
Message-ID: <DF145C04-4909-46BA-8412-9F1796269505@openstack.org>

Hi,

It is a friendly reminder that we are back to the US-Europe friendly time slot this week for the weekly call which is Tuesdays at 7am PT / 1500 UTC. We will start with a presentation about ETSI MEC.

For the full agenda and dial-in details please see the wiki: https://wiki.openstack.org/wiki/Edge_Computing_Group#Meetings

Thanks,
Ildikó




From bharath_ves at hotmail.com  Tue Feb 12 17:52:08 2019
From: bharath_ves at hotmail.com (bharath thiruveedula)
Date: Tue, 12 Feb 2019 17:52:08 +0000
Subject: [Edge-computing] Edge Node locator
Message-ID: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/0f2b9a33/attachment.html>

From srinivasa.r.addepalli at intel.com  Tue Feb 12 18:10:36 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Tue, 12 Feb 2019 18:10:36 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>

Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:


-        Latency/Distance with respect to UEs.

-        Availability of right hardware capabilities required for the workload.

-        Cost

-        Affinity and Anti-affinity rules

-        Etc...

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc...

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/6a7e8358/attachment.html>

From lbragstad at gmail.com  Tue Feb 12 18:25:24 2019
From: lbragstad at gmail.com (Lance Bragstad)
Date: Tue, 12 Feb 2019 12:25:24 -0600
Subject: [Edge-computing] [keystone] x509 authentication
In-Reply-To: <CAE6oFcEWjZ1i0f=gfTZyas6o=xoaH05KjRPFWiZyr1=FygwzvA@mail.gmail.com>
References: <CAE6oFcE=ODuMB_CVr_sz5Spd4Tn=EPcdYzsiqedDEf01xNParg@mail.gmail.com>
 <CAMomh-7TgzombrucdtW4idk88S2GORrBirNJwAGB3Vxp0JJsKQ@mail.gmail.com>
 <CAE6oFcEWjZ1i0f=gfTZyas6o=xoaH05KjRPFWiZyr1=FygwzvA@mail.gmail.com>
Message-ID: <de099fdf-9cf6-b799-0404-61af51c0a1eb@gmail.com>

Sending a quick update here that summarizes activity on this topic from
the last couple of weeks.

A few more bugs have trickled in regarding x509 federation support [0].
One of the original authors of the feature has started chipping away at
fixing them, but they can be worked in parallel if others are interested
in this work. As a reminder, there are areas of the docs that can be
improved, in case you don't have time to dig into a patch.

[0] https://bugs.launchpad.net/keystone/+bugs?field.tag=x509

On 1/29/19 11:55 AM, Lance Bragstad wrote:
>
>
> On Fri, Jan 25, 2019 at 3:02 PM James Penick <jpenick at gmail.com
> <mailto:jpenick at gmail.com>> wrote:
>
>     Hey Lance,
>      We'd definitely be interested in helping with the work. I'll grab
>     some volunteers from my team and get them in touch within the next
>     few days.
>
>
> Awesome, that sounds great! I'm open to using this thread for more
> technical communication if needed. Otherwise, #openstack-keystone is
> always open for folks to swing by if they want to discuss things there.
>
> FWIW - we brought this up in the keystone meeting today and there
> several other people interested in this work. There is probably going
> to be an opportunity to break the work up a bit.
>  
>
>     -James
>
>
>     On Fri, Jan 25, 2019 at 11:16 AM Lance Bragstad
>     <lbragstad at gmail.com <mailto:lbragstad at gmail.com>> wrote:
>
>         Hi all,
>
>         We've been going over keystone gaps that need to be addressed
>         for edge use cases every Tuesday. Since Berlin, Oath has
>         open-sourced some of their custom authentication plugins for
>         keystone that help them address these gaps.
>
>         The basic idea is that users authenticate to some external
>         identity provider (Athenz in Oath's case), and then present an
>         Athenz token to keystone. The custom plugins decode the token
>         from Athenz to determine the user, project, roles assignments,
>         and other useful bits of information. After that, it creates
>         any resources that don't exist in keystone already.
>         Ultimately, a user can authenticate against a keystone node
>         and have specific resources provisioned automatically. In
>         Berlin, engineers from Oath were saying they'd like to move
>         away from Athenz tokens altogether and use x509 certificates
>         issued by Athenz instead. The auto-provisioning approach is
>         very similar to a feature we have in keystone already. In
>         Berlin, and shortly after, there was general agreement that if
>         we could support x509 authentication with auto-provisioning
>         via keystone federation, that would pretty much solve Oath's
>         use case without having to maintain custom keystone plugins.
>
>         Last week, Colleen started digging into keystone's existing
>         x509 authentication support. I'll start with the good news,
>         which is x509 authentication works, for the most part. It's
>         been a feature in keystone for a long time, and it landed
>         after we implemented federation support around the Kilo
>         release. Chances are there won't be a need for a keystone
>         specification like we were initially thinking in the edge
>         meetings. Unfortunately, the implementation for x509
>         authentication has outdated documentation, is extremely
>         fragile, hard to set up, and hasn't been updated with
>         improvements we've made to the federation API since the
>         original implementation (like shadow users or
>         auto-provisioning, which work with other federated protocols
>         like OpenID Connect and SAML). We've started tracking the gaps
>         with bugs [0] so that we have things written down.
>
>         I think the good thing is that once we get this cleaned up,
>         we'll be able to re-use some of the newer federation features
>         with x509 authentication/federation. These updates would make
>         x509 a first-class federated protocol. The approach, pending
>         the bug fixes, would remove the need for Oath's custom
>         authentication plugins. It could be useful for edge
>         deployments, or even deployments with many regions, by
>         allowing users to be auto-provisioned in each region.
>         Although, it doesn't necessarily solve the network partition
>         issue.
>
>         Now that we have an idea of where to start and some bug
>         reports [0], I'm wondering if anyone is interested in helping
>         with the update or refactor. Because this won't require a
>         specification, we can get started on it sooner, instead of
>         having to wait for Train development and a new specification.
>         I'm also curious if anyone has comments or questions about the
>         approach.
>
>         Thanks,
>
>         Lance
>
>         [0] https://bugs.launchpad.net/keystone/+bugs?field.tag=x509
>         _______________________________________________
>         Edge-computing mailing list
>         Edge-computing at lists.openstack.org
>         <mailto:Edge-computing at lists.openstack.org>
>         http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/889800a9/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/889800a9/attachment-0001.sig>

From asayeed at redhat.com  Tue Feb 12 19:06:04 2019
From: asayeed at redhat.com (Azhar Sayeed)
Date: Tue, 12 Feb 2019 14:06:04 -0500
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
Message-ID: <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>

The way I have been thinking about this is the following..

You will have to separate infra functions such as the Edge Nodes from service functions such as applications that run on edge nodes and provide a service.

From an Infra management point of view..

Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those work loads based on certain criteria.
Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.

Both of them can be achieved via a service bus/message bus architecture that uses pub/sub concepts where you publish information and subscribe to topics. Edge nodes need to participate in that service bus model. It makes the discovery process and provisioning of applications to Edge nodes a lot easier and doesn’t require yet another discovery protocol… This is the work best suited for Orchestrators - So perhaps ONAP 

From a UE point of view - A UE should not care where an edge node is located. UE does a normal attach process to the network. The network maps the available services and shows them to UE via a catalog or another mechanism. The UE decides to either attach to that service or sign up for push notifications. So

A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the “closest” edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone” (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

Bottom line, A UE should not care where the Edge node is.

Regards,
Azhar

> On Feb 12, 2019, at 1:10 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com> wrote:
> 
> Hi Bharath,
>  
> JFYI
> This work is going on in ONAP (OOF project and Edge automation project).
>  
> Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).
>  
> ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:
>  
> -        Latency/Distance with respect to UEs.
> -        Availability of right hardware capabilities required for the workload.
> -        Cost
> -        Affinity and Anti-affinity rules
> -        Etc…
>  
> Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…
>  
> You may want to check out ONAP project.
>  
> Thanks
> Srini
>  
>  
>  
>  
> Thanks
> Srini
>  
>   <>
>  <>From: bharath thiruveedula [mailto:bharath_ves at hotmail.com] 
> Sent: Tuesday, February 12, 2019 9:52 AM
> To: edge-computing at lists.openstack.org
> Subject: [Edge-computing] Edge Node locator
>  
> Hi ,
>  
> Disclaimer: I am not sure the question following is in the scope of this forum.
>  
> I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?
>  
> Can you please share some pointers from this perspective?
>  
> Best Regards
> Bharath T
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/19040822/attachment.html>

From srinivasa.r.addepalli at intel.com  Tue Feb 12 19:29:20 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Tue, 12 Feb 2019 19:29:20 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026BA6@ORSMSX103.amr.corp.intel.com>

Yes, I agree with Azhar. UE APP should not know anything about edge location / node.

This is what was discussed in ONAP edge automation group.  This is one example flow.
In this example, UPF is used to redirect the traffic to right dynamic application. In simplest scenarios, this could be ADC or it could be that UEAPP is informed of IP address/FQDN of new dynamic application.

[cid:image002.jpg at 01D4C2C6.30B70180]

From: Azhar Sayeed [mailto:asayeed at redhat.com]
Sent: Tuesday, February 12, 2019 11:06 AM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

The way I have been thinking about this is the following..

You will have to separate infra functions such as the Edge Nodes from service functions such as applications that run on edge nodes and provide a service.

From an Infra management point of view..


  1.  Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those work loads based on certain criteria.
  2.  Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.

Both of them can be achieved via a service bus/message bus architecture that uses pub/sub concepts where you publish information and subscribe to topics. Edge nodes need to participate in that service bus model. It makes the discovery process and provisioning of applications to Edge nodes a lot easier and doesn’t require yet another discovery protocol… This is the work best suited for Orchestrators - So perhaps ONAP

From a UE point of view - A UE should not care where an edge node is located. UE does a normal attach process to the network. The network maps the available services and shows them to UE via a catalog or another mechanism. The UE decides to either attach to that service or sign up for push notifications. So


  1.  A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the “closest” edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
  2.  A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone” (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

Bottom line, A UE should not care where the Edge node is.

Regards,
Azhar

On Feb 12, 2019, at 1:10 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:

Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:

-        Latency/Distance with respect to UEs.
-        Availability of right hardware capabilities required for the workload.
-        Cost
-        Affinity and Anti-affinity rules
-        Etc…

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/21564d8a/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.jpg
Type: image/jpeg
Size: 88185 bytes
Desc: image002.jpg
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/21564d8a/attachment-0001.jpg>

From asayeed at redhat.com  Tue Feb 12 19:35:13 2019
From: asayeed at redhat.com (Azhar Sayeed)
Date: Tue, 12 Feb 2019 14:35:13 -0500
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026BA6@ORSMSX103.amr.corp.intel.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026BA6@ORSMSX103.amr.corp.intel.com>
Message-ID: <898803DE-0979-4473-A89C-0B79B24F8C18@redhat.com>

Yup!!… thanks Sreeni..thats what I was describing - you have a nice slide for it
Azhar

> On Feb 12, 2019, at 2:29 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com> wrote:
> 
> Yes, I agree with Azhar. UE APP should not know anything about edge location / node.
>  
> This is what was discussed in ONAP edge automation group.  This is one example flow.
> In this example, UPF is used to redirect the traffic to right dynamic application. In simplest scenarios, this could be ADC or it could be that UEAPP is informed of IP address/FQDN of new dynamic application.
>  
> <image002.jpg>
>   <>
>  <>From: Azhar Sayeed [mailto:asayeed at redhat.com] 
> Sent: Tuesday, February 12, 2019 11:06 AM
> To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
> Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
> Subject: Re: [Edge-computing] Edge Node locator
>  
> The way I have been thinking about this is the following..
>  
> You will have to separate infra functions such as the Edge Nodes from service functions such as applications that run on edge nodes and provide a service.
>  
> From an Infra management point of view..
>  
> Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those work loads based on certain criteria.
> Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.
>  
> Both of them can be achieved via a service bus/message bus architecture that uses pub/sub concepts where you publish information and subscribe to topics. Edge nodes need to participate in that service bus model. It makes the discovery process and provisioning of applications to Edge nodes a lot easier and doesn’t require yet another discovery protocol… This is the work best suited for Orchestrators - So perhaps ONAP 
>  
> From a UE point of view - A UE should not care where an edge node is located. UE does a normal attach process to the network. The network maps the available services and shows them to UE via a catalog or another mechanism. The UE decides to either attach to that service or sign up for push notifications. So
>  
> A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the “closest” edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
> A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone” (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models
>  
> Bottom line, A UE should not care where the Edge node is.
>  
> Regards,
> Azhar
>  
> On Feb 12, 2019, at 1:10 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com <mailto:srinivasa.r.addepalli at intel.com>> wrote:
>  
> Hi Bharath,
>  
> JFYI
> This work is going on in ONAP (OOF project and Edge automation project).
>  
> Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).
>  
> ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:
>  
> -        Latency/Distance with respect to UEs.
> -        Availability of right hardware capabilities required for the workload.
> -        Cost
> -        Affinity and Anti-affinity rules
> -        Etc…
>  
> Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…
>  
> You may want to check out ONAP project.
>  
> Thanks
> Srini
>  
>  
>  
>  
> Thanks
> Srini
>  
>  
> From: bharath thiruveedula [mailto:bharath_ves at hotmail.com <mailto:bharath_ves at hotmail.com>] 
> Sent: Tuesday, February 12, 2019 9:52 AM
> To: edge-computing at lists.openstack.org <mailto:edge-computing at lists.openstack.org>
> Subject: [Edge-computing] Edge Node locator
>  
> Hi ,
>  
> Disclaimer: I am not sure the question following is in the scope of this forum.
>  
> I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?
>  
> Can you please share some pointers from this perspective?
>  
> Best Regards
> Bharath T
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org <mailto:Edge-computing at lists.openstack.org>
> http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing <http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/26fb53b1/attachment-0001.html>

From rob at rackn.com  Tue Feb 12 20:22:33 2019
From: rob at rackn.com (Rob Hirschfeld)
Date: Tue, 12 Feb 2019 14:22:33 -0600
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
Message-ID: <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>

This needs to be corrected ASAP since they are both spotlighted as LF Edge
projects > "MobileEdgeX and LFE EVE projects (These two are not yet open
sourced)."

Rob Hirschfeld





On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <
srinivasa.r.addepalli at intel.com> wrote:

> Hi Bharath,
>
>
>
> JFYI
>
> This work is going on in ONAP (OOF project and Edge automation project).
>
>
>
> Other possible projects are  MobileEdgeX and LFE EVE projects (These two
> are not yet open sourced).
>
>
>
> ONAP OOF project intention is to figure out the best edge location for a
> given workload based on following criteria:
>
>
>
> -        Latency/Distance with respect to UEs.
>
> -        Availability of right hardware capabilities required for the
> workload.
>
> -        Cost
>
> -        Affinity and Anti-affinity rules
>
> -        Etc…
>
>
>
> Once ONAP OOF chooses the edge location to  place the workload, then ONAP
> (via Multi-Cloud project) will place the workload by calling Openstack API
> or K8S API etc…
>
>
>
> You may want to check out ONAP project.
>
>
>
> Thanks
>
> Srini
>
>
>
>
>
>
>
>
>
> Thanks
>
> Srini
>
>
>
>
>
> *From:* bharath thiruveedula [mailto:bharath_ves at hotmail.com]
> *Sent:* Tuesday, February 12, 2019 9:52 AM
> *To:* edge-computing at lists.openstack.org
> *Subject:* [Edge-computing] Edge Node locator
>
>
>
> Hi ,
>
>
>
> Disclaimer: I am not sure the question following is in the scope of this
> forum.
>
>
>
> I am trying to understand the Edge Architecture from UE point of view. How
> an app in UE can detect what the nearest edge it has to contact and how
> does it know whether the app is provisioned in that edge or not?
>
>
>
> Can you please share some pointers from this perspective?
>
>
>
> Best Regards
>
> Bharath T
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/7fcc1815/attachment.html>

From srinivasa.r.addepalli at intel.com  Tue Feb 12 20:47:55 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Tue, 12 Feb 2019 20:47:55 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>

MobileEdgeX is not yet part of LFE.  It may that you might be referring to EdgeXFoundry.  MobileEdgX and EdgeXFoundry are two different projects addressing two different scenarios.

Thanks
Srini


From: Rob Hirschfeld [mailto:rob at rackn.com]
Sent: Tuesday, February 12, 2019 12:23 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

This needs to be corrected ASAP since they are both spotlighted as LF Edge projects > "MobileEdgeX and LFE EVE projects (These two are not yet open sourced)."



Rob Hirschfeld






On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:
Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:


-        Latency/Distance with respect to UEs.

-        Availability of right hardware capabilities required for the workload.

-        Cost

-        Affinity and Anti-affinity rules

-        Etc…

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/fa9faf75/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 332 bytes
Desc: image001.jpg
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/fa9faf75/attachment-0001.jpg>

From rob at rackn.com  Tue Feb 12 20:50:36 2019
From: rob at rackn.com (Rob Hirschfeld)
Date: Tue, 12 Feb 2019 14:50:36 -0600
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>
Message-ID: <CAG9nXnhPKa2fEZhPg9S98UMyavMW25KYBkzCOkroH3d_bN0uXw@mail.gmail.com>

Thank you for clarifying.  I suspect I'm not the only person getting
confused about that.

<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Fwww.rackn.com%2F&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>

Rob Hirschfeld

Co-Founder and Chief Executive Officer, RackN

(512) 773-7522

<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Ftwitter.com%2Fzehicle%3Flang%3Den&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>
<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Fwww.linkedin.com%2Fin%2Frhirschfeld%2F&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>



On Tue, Feb 12, 2019 at 2:47 PM Addepalli, Srinivasa R <
srinivasa.r.addepalli at intel.com> wrote:

> MobileEdgeX is not yet part of LFE.  It may that you might be referring to
> EdgeXFoundry.  MobileEdgX and EdgeXFoundry are two different projects
> addressing two different scenarios.
>
>
>
> Thanks
>
> Srini
>
>
>
>
>
> *From:* Rob Hirschfeld [mailto:rob at rackn.com]
> *Sent:* Tuesday, February 12, 2019 12:23 PM
> *To:* Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
> *Cc:* bharath thiruveedula <bharath_ves at hotmail.com>;
> edge-computing at lists.openstack.org; ramkik at vmware.com
> *Subject:* Re: [Edge-computing] Edge Node locator
>
>
>
> This needs to be corrected ASAP since they are both spotlighted as LF Edge
> projects > "MobileEdgeX and LFE EVE projects (These two are not yet open
> sourced)."
>
>
>
> *Rob Hirschfeld*
>
>
>
>
> [image: Image removed by sender.]
>
>
>
> On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <
> srinivasa.r.addepalli at intel.com> wrote:
>
> Hi Bharath,
>
>
>
> JFYI
>
> This work is going on in ONAP (OOF project and Edge automation project).
>
>
>
> Other possible projects are  MobileEdgeX and LFE EVE projects (These two
> are not yet open sourced).
>
>
>
> ONAP OOF project intention is to figure out the best edge location for a
> given workload based on following criteria:
>
>
>
> -        Latency/Distance with respect to UEs.
>
> -        Availability of right hardware capabilities required for the
> workload.
>
> -        Cost
>
> -        Affinity and Anti-affinity rules
>
> -        Etc…
>
>
>
> Once ONAP OOF chooses the edge location to  place the workload, then ONAP
> (via Multi-Cloud project) will place the workload by calling Openstack API
> or K8S API etc…
>
>
>
> You may want to check out ONAP project.
>
>
>
> Thanks
>
> Srini
>
>
>
>
>
>
>
>
>
> Thanks
>
> Srini
>
>
>
>
>
> *From:* bharath thiruveedula [mailto:bharath_ves at hotmail.com]
> *Sent:* Tuesday, February 12, 2019 9:52 AM
> *To:* edge-computing at lists.openstack.org
> *Subject:* [Edge-computing] Edge Node locator
>
>
>
> Hi ,
>
>
>
> Disclaimer: I am not sure the question following is in the scope of this
> forum.
>
>
>
> I am trying to understand the Edge Architecture from UE point of view. How
> an app in UE can detect what the nearest edge it has to contact and how
> does it know whether the app is provisioned in that edge or not?
>
>
>
> Can you please share some pointers from this perspective?
>
>
>
> Best Regards
>
> Bharath T
>
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/6e75fa60/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 332 bytes
Desc: not available
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/6e75fa60/attachment.jpg>

From srinivasa.r.addepalli at intel.com  Tue Feb 12 20:59:55 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Tue, 12 Feb 2019 20:59:55 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CAG9nXnhPKa2fEZhPg9S98UMyavMW25KYBkzCOkroH3d_bN0uXw@mail.gmail.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhPKa2fEZhPg9S98UMyavMW25KYBkzCOkroH3d_bN0uXw@mail.gmail.com>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026FE2@ORSMSX103.amr.corp.intel.com>

Yes. It is confusing due to common term ‘EdgeX’ in both places.

JFYI

EdgeXFoundry : Is IOT platform.
MobileEdgeX, EVE and ONAP:  Mainly to orchestrate and automate applications/VNFs on several edge locations.

There is  confusion with Airship too.  Airship is for bare-metal provisioning with respect to deploying host operating system and related utilities/processes to make edge/DC ready.

Thanks
Srini


From: Rob Hirschfeld [mailto:rob at rackn.com]
Sent: Tuesday, February 12, 2019 12:51 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

Thank you for clarifying.  I suspect I'm not the only person getting confused about that.


[https://lh6.googleusercontent.com/ySrnpZRlg0FK6BwnZw-GLxA8PuamwCARuiMJ_VjL4muhF_ifki9TWNgZvtavVZKSv_2mEBFYbr6BEeeb6BkBiJ0l27qL7yCHC7hsPvE1ZbTkyY30JPVO_uKHlH2w3yv4OQI-WUyg]<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Fwww.rackn.com%2F&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>


Rob Hirschfeld

Co-Founder and Chief Executive Officer, RackN

(512) 773-7522


[https://lh6.googleusercontent.com/HlA5qvRA4ba1F5C0kOubBXP0qF4DlXlIdKyQJG_sWo2FKAIZbsicvGS-8F9WbXQ2ZNLySHg-bb1W3V2BxSR1Z8vpd7mOH798iSxdmvo8q75DqVkq29Cj7Y5VvqiB9sViJwHqSSNY]<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Ftwitter.com%2Fzehicle%3Flang%3Den&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>[https://lh6.googleusercontent.com/fQXQPBYLS6cl3yZhjq2LhPlzqrCc5WLLZcdFqxsSQjis3xb34H5hoaZkcAYpHIgKEksvW6qhrZYFCscWitnWmR3yTJ34uLfTX0ozwOl_JwJn9c1VFpTtq3xj15SNw66zt8X3-rHc]<https://t.sidekickopen10.com/s2t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02?t=https%3A%2F%2Fwww.linkedin.com%2Fin%2Frhirschfeld%2F&si=5660451956457472&pi=e084781c-677c-403c-d2c8-81edf582fdae>





On Tue, Feb 12, 2019 at 2:47 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:
MobileEdgeX is not yet part of LFE.  It may that you might be referring to EdgeXFoundry.  MobileEdgX and EdgeXFoundry are two different projects addressing two different scenarios.

Thanks
Srini


From: Rob Hirschfeld [mailto:rob at rackn.com<mailto:rob at rackn.com>]
Sent: Tuesday, February 12, 2019 12:23 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>>
Cc: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator

This needs to be corrected ASAP since they are both spotlighted as LF Edge projects > "MobileEdgeX and LFE EVE projects (These two are not yet open sourced)."



Rob Hirschfeld






On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:
Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:


-        Latency/Distance with respect to UEs.

-        Availability of right hardware capabilities required for the workload.

-        Cost

-        Affinity and Anti-affinity rules

-        Etc…

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/080bf955/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 332 bytes
Desc: image001.jpg
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/080bf955/attachment-0001.jpg>

From Sebastian.Robitzsch at InterDigital.com  Tue Feb 12 23:06:53 2019
From: Sebastian.Robitzsch at InterDigital.com (Sebastian Robitzsch)
Date: Tue, 12 Feb 2019 23:06:53 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B026FE2@ORSMSX103.amr.corp.intel.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhPKa2fEZhPg9S98UMyavMW25KYBkzCOkroH3d_bN0uXw@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026FE2@ORSMSX103.amr.corp.intel.com>
Message-ID: <BN6PR1001MB2081A9974A568FD3D972732791650@BN6PR1001MB2081.namprd10.prod.outlook.com>

Dear all,

Let me chime in at this point and go back to the very interesting question Bharat asked:  “am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?”

He asked about architectural viewpoints and not solutions which were mainly presented as an answer so far. From a system’s perspective the UE is/will eventually become part of the networking system with the edge expanding to the terminal (leaving it up to your interpretation what a terminal could be – ranging from traditional handsets, to all-in-one devices up to constraint devices that have only particular functionalities (such as displaying content, processing them or storing them)).
In all UE/terminal-centric views the only task an endpoint must solve is how it can communicate with the system (point of attachment) in order to become part of it. Next is to announce/negotiate the resources so that its cloud resources (CPU, RAM, disk, networking) can be orchestrated correctly by the infrastructure’s orchestrator.
If the endpoint is seeking other resources inside/outside the system’s domain (such as an FQDN in the modern world) a resource pool is queried (traditionally DNS) which results in a communication between the initializing endpoint and the serving endpoint – across domains.

An edge compute architecture (from OpenStack’s aka Open Infrastructure’s point of view) allows the orchestration of infrastructure resources which might include the terminal. But it should leave it to the system / platform deployed on top of the infrastructure to determine the where a service initializer on the UE (application) is receiving its information from; this could be a traditional IP stack with DNS (not very dynamic and scalable in a modern edge environment) or a more edge-focused solution which does not necessarily follow the traditional way the internet works (see ETSI MEC PoC’s).

Bottom line, the app in the UE does not “detect” the nearest edge. In an ideal world the routing inside the network system is fully decoupled from the information two or more endpoints would like to exchange (e.g. your app on the UE and a service in some (edge) instance). The shortcomings of DNS and how services are tight to an IP address which immediately binds you physically to an instance where the service is instantiated is well understood as a problem. Several solutions have various answers to that (floating IPs, customised DNS black-boxes as VNFs, triangular routing, tunneling, etc). But one thing I’d like to stress: the UE must not detect the nearest edge itself. There’s approaches like service function chaining and name-based routing where any application simply communicates its intent what service to access to the platform deployed on top of the infrastructure. How the actual request makes it to the service instance is decoupled from the intend of the application on the UE. And this is what an edge (infrastructure) architecture should support and outline in its specification.


Best Regards,
Sebastian

From: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Sent: 12 February 2019 21:00
To: Rob Hirschfeld <rob at rackn.com>
Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

Yes. It is confusing due to common term ‘EdgeX’ in both places.

JFYI

EdgeXFoundry : Is IOT platform.
MobileEdgeX, EVE and ONAP:  Mainly to orchestrate and automate applications/VNFs on several edge locations.

There is  confusion with Airship too.  Airship is for bare-metal provisioning with respect to deploying host operating system and related utilities/processes to make edge/DC ready.

Thanks
Srini


From: Rob Hirschfeld [mailto:rob at rackn.com]
Sent: Tuesday, February 12, 2019 12:51 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>>
Cc: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator

Thank you for clarifying.  I suspect I'm not the only person getting confused about that.


[https://lh6.googleusercontent.com/ySrnpZRlg0FK6BwnZw-GLxA8PuamwCARuiMJ_VjL4muhF_ifki9TWNgZvtavVZKSv_2mEBFYbr6BEeeb6BkBiJ0l27qL7yCHC7hsPvE1ZbTkyY30JPVO_uKHlH2w3yv4OQI-WUyg]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Fwww.rackn.com%252F%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517650301&sdata=0oOno9RQ%2FpJhCh%2BB0IVu3VIX9DyVUSET6rZoCpGhFRw%3D&reserved=0>


Rob Hirschfeld

Co-Founder and Chief Executive Officer, RackN

(512) 773-7522


[https://lh6.googleusercontent.com/HlA5qvRA4ba1F5C0kOubBXP0qF4DlXlIdKyQJG_sWo2FKAIZbsicvGS-8F9WbXQ2ZNLySHg-bb1W3V2BxSR1Z8vpd7mOH798iSxdmvo8q75DqVkq29Cj7Y5VvqiB9sViJwHqSSNY]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Ftwitter.com%252Fzehicle%253Flang%253Den%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517650301&sdata=ltzE1WmpjGHJaq0dSxxzZKb3N31oQ9l28Y11iKon8QI%3D&reserved=0>[https://lh6.googleusercontent.com/fQXQPBYLS6cl3yZhjq2LhPlzqrCc5WLLZcdFqxsSQjis3xb34H5hoaZkcAYpHIgKEksvW6qhrZYFCscWitnWmR3yTJ34uLfTX0ozwOl_JwJn9c1VFpTtq3xj15SNw66zt8X3-rHc]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Fwww.linkedin.com%252Fin%252Frhirschfeld%252F%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517660306&sdata=GRoJ4LWkUQ%2B7H898HX%2FVz%2Bmlj5FZDiC5fGSPaMwRhBU%3D&reserved=0>





On Tue, Feb 12, 2019 at 2:47 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:
MobileEdgeX is not yet part of LFE.  It may that you might be referring to EdgeXFoundry.  MobileEdgX and EdgeXFoundry are two different projects addressing two different scenarios.

Thanks
Srini


From: Rob Hirschfeld [mailto:rob at rackn.com<mailto:rob at rackn.com>]
Sent: Tuesday, February 12, 2019 12:23 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>>
Cc: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator

This needs to be corrected ASAP since they are both spotlighted as LF Edge projects > "MobileEdgeX and LFE EVE projects (These two are not yet open sourced)."



Rob Hirschfeld






On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:
Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:


-        Latency/Distance with respect to UEs.

-        Availability of right hardware capabilities required for the workload.

-        Cost

-        Affinity and Anti-affinity rules

-        Etc…

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.openstack.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fedge-computing&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517670316&sdata=n9uveD%2BZidGfWr0SUDBN2ZyR1d63Cils1iqftUbO1os%3D&reserved=0>
[Banner]


[Banner] [Banner]
This e-mail is intended only for the use of the individual or entity to which it is addressed, and may contain information that is privileged, confidential and/or otherwise protected from disclosure to anyone other than its intended recipient. Unintended transmission shall not constitute waiver of any privilege or confidentiality obligation. If you received this communication in error, please do not review, copy or distribute it, notify me immediately by email, and delete the original message and any attachments. Unless expressly stated in this e-mail, nothing in this message or any attachment should be construed as a digital or electronic signature.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/3c53e8ce/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 332 bytes
Desc: image001.jpg
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/3c53e8ce/attachment-0001.jpg>

From bharath_ves at hotmail.com  Wed Feb 13 14:07:37 2019
From: bharath_ves at hotmail.com (bharath thiruveedula)
Date: Wed, 13 Feb 2019 14:07:37 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>,
 <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
Message-ID: <BMXPR01MB2278D30C92EF3725D12FB6AA94660@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>

Hi Srini/Sayeed,

Thanks a lot for the interesting points

If not  UE directly detecting the edge node, I was looking for the solution where the service request from UE will be redirected to the edge.


  1.  Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those  based on certain criteria.
  2.  Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.

I think these will hold good, if edge is installed/deployed and configured to communicate  to Edge Orchestrator. We can also provision the edge from the Orchestrator itself. In that case, we don't need advertise and discover the Edge node and it will be part of the inventory. But I think the former case is the better option.


  1.  A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the “closest” edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
  2.  A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone” (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

For this process, we need changes in the UE I believe. Is there any documentation for more details?

Srini, I was looking for similar solutions in ONAP. The work you mentioned in OOF will be part of Dublin? Are you referring to FGPS?

Best Regards
Bharath T

________________________________
From: Azhar Sayeed <asayeed at redhat.com>
Sent: Wednesday, February 13, 2019 12:36 AM
To: Addepalli, Srinivasa R
Cc: bharath thiruveedula; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

The way I have been thinking about this is the following..

You will have to separate infra functions such as the Edge Nodes from service functions such as applications that run on edge nodes and provide a service.

>From an Infra management point of view..


  1.  Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those work loads based on certain criteria.
  2.  Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.

Both of them can be achieved via a service bus/message bus architecture that uses pub/sub concepts where you publish information and subscribe to topics. Edge nodes need to participate in that service bus model. It makes the discovery process and provisioning of applications to Edge nodes a lot easier and doesn’t require yet another discovery protocol… This is the work best suited for Orchestrators - So perhaps ONAP

>From a UE point of view - A UE should not care where an edge node is located. UE does a normal attach process to the network. The network maps the available services and shows them to UE via a catalog or another mechanism. The UE decides to either attach to that service or sign up for push notifications. So


  1.  A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the “closest” edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
  2.  A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone” (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

Bottom line, A UE should not care where the Edge node is.

Regards,
Azhar

On Feb 12, 2019, at 1:10 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:

Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:

-        Latency/Distance with respect to UEs.
-        Availability of right hardware capabilities required for the workload.
-        Cost
-        Affinity and Anti-affinity rules
-        Etc…

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/b442eb44/attachment-0001.html>

From christopher.price at est.tech  Wed Feb 13 14:17:05 2019
From: christopher.price at est.tech (Christopher Price)
Date: Wed, 13 Feb 2019 14:17:05 +0000
Subject: [Edge-computing] Edge Node locator
Message-ID: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>

Bharath,

A left field thought which I find quite interesting is to run your application on a CDN overlay.
This can be constrained to the application layer and run over IP which reduces the need to try and weld this type of function to the infrastructure layer or expose too much service awareness through an extended mesh.  (security security)  Moves the problem to a defined space and keeps your infrastructure and app workloads simple and focused on what they should be doing.

Millions of ways to skin that poor little cat though, imagination Vs cost…

/ Chris

From: bharath thiruveedula <bharath_ves at hotmail.com>
Date: Tuesday, 12 February 2019 at 18:53
To: "edge-computing at lists.openstack.org" <edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/8ea4ad12/attachment.html>

From christopher.price at est.tech  Wed Feb 13 14:20:11 2019
From: christopher.price at est.tech (Christopher Price)
Date: Wed, 13 Feb 2019 14:20:11 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
References: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
Message-ID: <0C574EFD-B6A4-44BB-96A5-449053B7DC80@est.tech>

Sorry I was “auto-corrected”:  CBN (content based networking), not CDN.

From: Christopher Price <christopher.price at est.tech>
Date: Wednesday, 13 February 2019 at 15:17
To: bharath thiruveedula <bharath_ves at hotmail.com>, "edge-computing at lists.openstack.org" <edge-computing at lists.openstack.org>
Subject: Re: [Edge-computing] Edge Node locator

Bharath,

A left field thought which I find quite interesting is to run your application on a CDN overlay.
This can be constrained to the application layer and run over IP which reduces the need to try and weld this type of function to the infrastructure layer or expose too much service awareness through an extended mesh.  (security security)  Moves the problem to a defined space and keeps your infrastructure and app workloads simple and focused on what they should be doing.

Millions of ways to skin that poor little cat though, imagination Vs cost…

/ Chris

From: bharath thiruveedula <bharath_ves at hotmail.com>
Date: Tuesday, 12 February 2019 at 18:53
To: "edge-computing at lists.openstack.org" <edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/d5ddb68d/attachment-0001.html>

From bharath_ves at hotmail.com  Wed Feb 13 14:20:39 2019
From: bharath_ves at hotmail.com (bharath thiruveedula)
Date: Wed, 13 Feb 2019 14:20:39 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <BN6PR1001MB2081A9974A568FD3D972732791650@BN6PR1001MB2081.namprd10.prod.outlook.com>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhtrtDdW8axb5t_mQ8rSBYQkvboUb8zo_sJG9psL6KwBA@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026EAD@ORSMSX103.amr.corp.intel.com>
 <CAG9nXnhPKa2fEZhPg9S98UMyavMW25KYBkzCOkroH3d_bN0uXw@mail.gmail.com>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B026FE2@ORSMSX103.amr.corp.intel.com>,
 <BN6PR1001MB2081A9974A568FD3D972732791650@BN6PR1001MB2081.namprd10.prod.outlook.com>
Message-ID: <BMXPR01MB22785566D0BEA1A087D2BF3394660@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>

Hi Sebastian,

Yeah I was actually looking for architecture point of view rather than actual solutions.

>> From a system’s perspective the UE is/will eventually become part of the networking system

Exactly, I was looking for such discussions in edge computing forum. In the current demos of MEC, are we considering the UE with such functionalities?

>> The shortcomings of DNS and how services are tight to an IP address which immediately binds you physically to an instance where the service is instantiated is well understood as a problem. Several solutions have various answers to that (floating IPs,  DNS black-boxes as VNFs, triangular routing, tunneling, etc)

This is interesting. How can the IP be resolved if the UE is moving and the DNS has to be refreshed if the edge is changed that serving the application? I am not sure, if it is already discussed.

Best Regards
Bharath T

________________________________
From: Sebastian Robitzsch <Sebastian.Robitzsch at InterDigital.com>
Sent: Wednesday, February 13, 2019 4:36 AM
To: Addepalli, Srinivasa R; Rob Hirschfeld
Cc: bharath thiruveedula; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: RE: [Edge-computing] Edge Node locator


Dear all,



Let me chime in at this point and go back to the very interesting question Bharat asked:  “am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?”



He asked about architectural viewpoints and not solutions which were mainly presented as an answer so far. From a system’s perspective the UE is/will eventually become part of the networking system with the edge expanding to the terminal (leaving it up to your interpretation what a terminal could be – ranging from traditional handsets, to all-in-one devices up to constraint devices that have only particular functionalities (such as displaying content, processing them or storing them)).

In all UE/terminal-centric views the only task an endpoint must solve is how it can communicate with the system (point of attachment) in order to become part of it. Next is to announce/negotiate the resources so that its cloud resources (CPU, RAM, disk, networking) can be orchestrated correctly by the infrastructure’s orchestrator.

If the endpoint is seeking other resources inside/outside the system’s domain (such as an FQDN in the modern world) a resource pool is queried (traditionally DNS) which results in a communication between the initializing endpoint and the serving endpoint – across domains.



An edge compute architecture (from OpenStack’s aka Open Infrastructure’s point of view) allows the orchestration of infrastructure resources which might include the terminal. But it should leave it to the system / platform deployed on top of the infrastructure to determine the where a service initializer on the UE (application) is receiving its information from; this could be a traditional IP stack with DNS (not very dynamic and scalable in a modern edge environment) or a more edge-focused solution which does not necessarily follow the traditional way the internet works (see ETSI MEC PoC’s).



Bottom line, the app in the UE does not “detect” the nearest edge. In an ideal world the routing inside the network system is fully decoupled from the information two or more endpoints would like to exchange (e.g. your app on the UE and a service in some (edge) instance). The shortcomings of DNS and how services are tight to an IP address which immediately binds you physically to an instance where the service is instantiated is well understood as a problem. Several solutions have various answers to that (floating IPs, customised DNS black-boxes as VNFs, triangular routing, tunneling, etc). But one thing I’d like to stress: the UE must not detect the nearest edge itself. There’s approaches like service function chaining and name-based routing where any application simply communicates its intent what service to access to the platform deployed on top of the infrastructure. How the actual request makes it to the service instance is decoupled from the intend of the application on the UE. And this is what an edge (infrastructure) architecture should support and outline in its specification.





Best Regards,

Sebastian



From: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Sent: 12 February 2019 21:00
To: Rob Hirschfeld <rob at rackn.com>
Cc: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator



Yes. It is confusing due to common term ‘EdgeX’ in both places.



JFYI



EdgeXFoundry : Is IOT platform.

MobileEdgeX, EVE and ONAP:  Mainly to orchestrate and automate applications/VNFs on several edge locations.



There is  confusion with Airship too.  Airship is for bare-metal provisioning with respect to deploying host operating system and related utilities/processes to make edge/DC ready.



Thanks

Srini





From: Rob Hirschfeld [mailto:rob at rackn.com]
Sent: Tuesday, February 12, 2019 12:51 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>>
Cc: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator



Thank you for clarifying.  I suspect I'm not the only person getting confused about that.



[https://lh6.googleusercontent.com/ySrnpZRlg0FK6BwnZw-GLxA8PuamwCARuiMJ_VjL4muhF_ifki9TWNgZvtavVZKSv_2mEBFYbr6BEeeb6BkBiJ0l27qL7yCHC7hsPvE1ZbTkyY30JPVO_uKHlH2w3yv4OQI-WUyg]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Fwww.rackn.com%252F%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517650301&sdata=0oOno9RQ%2FpJhCh%2BB0IVu3VIX9DyVUSET6rZoCpGhFRw%3D&reserved=0>


Rob Hirschfeld

Co-Founder and Chief Executive Officer, RackN

(512) 773-7522


[https://lh6.googleusercontent.com/HlA5qvRA4ba1F5C0kOubBXP0qF4DlXlIdKyQJG_sWo2FKAIZbsicvGS-8F9WbXQ2ZNLySHg-bb1W3V2BxSR1Z8vpd7mOH798iSxdmvo8q75DqVkq29Cj7Y5VvqiB9sViJwHqSSNY]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Ftwitter.com%252Fzehicle%253Flang%253Den%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517650301&sdata=ltzE1WmpjGHJaq0dSxxzZKb3N31oQ9l28Y11iKon8QI%3D&reserved=0>[https://lh6.googleusercontent.com/fQXQPBYLS6cl3yZhjq2LhPlzqrCc5WLLZcdFqxsSQjis3xb34H5hoaZkcAYpHIgKEksvW6qhrZYFCscWitnWmR3yTJ34uLfTX0ozwOl_JwJn9c1VFpTtq3xj15SNw66zt8X3-rHc]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Ft.sidekickopen10.com%2Fs2t%2Fc%2F5%2Ff18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XZs3M2bjgW64zKpj3LyR22W1qMCsg56dVpDf9jJGjx02%3Ft%3Dhttps%253A%252F%252Fwww.linkedin.com%252Fin%252Frhirschfeld%252F%26si%3D5660451956457472%26pi%3De084781c-677c-403c-d2c8-81edf582fdae&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517660306&sdata=GRoJ4LWkUQ%2B7H898HX%2FVz%2Bmlj5FZDiC5fGSPaMwRhBU%3D&reserved=0>








On Tue, Feb 12, 2019 at 2:47 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:

MobileEdgeX is not yet part of LFE.  It may that you might be referring to EdgeXFoundry.  MobileEdgX and EdgeXFoundry are two different projects addressing two different scenarios.



Thanks

Srini





From: Rob Hirschfeld [mailto:rob at rackn.com<mailto:rob at rackn.com>]
Sent: Tuesday, February 12, 2019 12:23 PM
To: Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>>
Cc: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator



This needs to be corrected ASAP since they are both spotlighted as LF Edge projects > "MobileEdgeX and LFE EVE projects (These two are not yet open sourced)."




Rob Hirschfeld









On Tue, Feb 12, 2019 at 12:12 PM Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:

Hi Bharath,



JFYI

This work is going on in ONAP (OOF project and Edge automation project).



Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).



ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:



-        Latency/Distance with respect to UEs.

-        Availability of right hardware capabilities required for the workload.

-        Cost

-        Affinity and Anti-affinity rules

-        Etc…



Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc…



You may want to check out ONAP project.



Thanks

Srini









Thanks

Srini





From: bharath thiruveedula [mailto:bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator



Hi ,



Disclaimer: I am not sure the question following is in the scope of this forum.



I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?



Can you please share some pointers from this perspective?



Best Regards

Bharath T

_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.openstack.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fedge-computing&data=02%7C01%7Csebastian.robitzsch%40interdigital.com%7Ce967d11039a349c2deb208d6912dde14%7Ce351b779f6d54e50856880e922d180ae%7C0%7C0%7C636856023517670316&sdata=n9uveD%2BZidGfWr0SUDBN2ZyR1d63Cils1iqftUbO1os%3D&reserved=0>

[Banner]


[Banner] [Banner]
This e-mail is intended only for the use of the individual or entity to which it is addressed, and may contain information that is privileged, confidential and/or otherwise protected from disclosure to anyone other than its intended recipient. Unintended transmission shall not constitute waiver of any privilege or confidentiality obligation. If you received this communication in error, please do not review, copy or distribute it, notify me immediately by email, and delete the original message and any attachments. Unless expressly stated in this e-mail, nothing in this message or any attachment should be construed as a digital or electronic signature.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/dc139c44/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 332 bytes
Desc: image001.jpg
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/dc139c44/attachment-0001.jpg>

From bharath_ves at hotmail.com  Wed Feb 13 14:28:23 2019
From: bharath_ves at hotmail.com (bharath thiruveedula)
Date: Wed, 13 Feb 2019 14:28:23 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
References: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
Message-ID: <BMXPR01MB2278254959CF296DBDA6239B94660@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>

Hi Christopher,

I was not aware of CBN. This seems very interesting. I will look into it for more details.

>> Millions of ways to skin that poor little cat though, imagination Vs cost…

That's true ! :)

Best Regards
Bharath T
________________________________
From: Christopher Price <christopher.price at est.tech>
Sent: Wednesday, February 13, 2019 7:47 PM
To: bharath thiruveedula; edge-computing at lists.openstack.org
Subject: Re: [Edge-computing] Edge Node locator


Bharath,



A left field thought which I find quite interesting is to run your application on a CDN overlay.

This can be constrained to the application layer and run over IP which reduces the need to try and weld this type of function to the infrastructure layer or expose too much service awareness through an extended mesh.  (security security)  Moves the problem to a defined space and keeps your infrastructure and app workloads simple and focused on what they should be doing.



Millions of ways to skin that poor little cat though, imagination Vs cost…



/ Chris



From: bharath thiruveedula <bharath_ves at hotmail.com>
Date: Tuesday, 12 February 2019 at 18:53
To: "edge-computing at lists.openstack.org" <edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator



Hi ,



Disclaimer: I am not sure the question following is in the scope of this forum.



I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?



Can you please share some pointers from this perspective?



Best Regards

Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/cfc2df77/attachment.html>

From srinivasa.r.addepalli at intel.com  Wed Feb 13 14:54:19 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Wed, 13 Feb 2019 14:54:19 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <0C574EFD-B6A4-44BB-96A5-449053B7DC80@est.tech>
References: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
 <0C574EFD-B6A4-44BB-96A5-449053B7DC80@est.tech>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B028BC3@ORSMSX103.amr.corp.intel.com>

It sounds like “Named Data Networking”.   If so, I agree, this is one of the novel approaches to name content object as the endpoint instead of IP address.  I think it is still in research phase and not sure whether it is adopted by endpoints and network infrastructure yet.  Any deployments or plans you are aware of?

Srini


From: Christopher Price [mailto:christopher.price at est.tech]
Sent: Wednesday, February 13, 2019 6:20 AM
To: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org
Subject: Re: [Edge-computing] Edge Node locator

Sorry I was “auto-corrected”:  CBN (content based networking), not CDN.

From: Christopher Price <christopher.price at est.tech<mailto:christopher.price at est.tech>>
Date: Wednesday, 13 February 2019 at 15:17
To: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>, "edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>" <edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>>
Subject: Re: [Edge-computing] Edge Node locator

Bharath,

A left field thought which I find quite interesting is to run your application on a CDN overlay.
This can be constrained to the application layer and run over IP which reduces the need to try and weld this type of function to the infrastructure layer or expose too much service awareness through an extended mesh.  (security security)  Moves the problem to a defined space and keeps your infrastructure and app workloads simple and focused on what they should be doing.

Millions of ways to skin that poor little cat though, imagination Vs cost…

/ Chris

From: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>
Date: Tuesday, 12 February 2019 at 18:53
To: "edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>" <edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/eb338ae1/attachment-0001.html>

From christopher.price at est.tech  Wed Feb 13 15:01:29 2019
From: christopher.price at est.tech (Christopher Price)
Date: Wed, 13 Feb 2019 15:01:29 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <CF008BD37B7B9A4BA7FD12ECEC3C98016B028BC3@ORSMSX103.amr.corp.intel.com>
References: <C934C860-1883-4D4F-BC35-62B6D651C449@est.tech>
 <0C574EFD-B6A4-44BB-96A5-449053B7DC80@est.tech>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B028BC3@ORSMSX103.amr.corp.intel.com>
Message-ID: <4671CFAB-505F-4BEF-AE1B-DE618E6AFA6F@est.tech>

Yes it’s been under investigation/development for a long long time. 😊
Supported in routers as of today, however you need an application layer that takes advantage of it and well let’s be honest IP prevails in a centralized architecture.  We seem as an industry to be moving toward a scenario where these types of approaches may begin to become viable, I’m not sure if it would really take off as anything other than an overlay but as SDN approaches the edge...
/ Chris

From: "Addepalli, Srinivasa R" <srinivasa.r.addepalli at intel.com>
Date: Wednesday, 13 February 2019 at 15:54
To: Christopher Price <christopher.price at est.tech>, bharath thiruveedula <bharath_ves at hotmail.com>, "edge-computing at lists.openstack.org" <edge-computing at lists.openstack.org>
Subject: RE: [Edge-computing] Edge Node locator

It sounds like “Named Data Networking”.   If so, I agree, this is one of the novel approaches to name content object as the endpoint instead of IP address.  I think it is still in research phase and not sure whether it is adopted by endpoints and network infrastructure yet.  Any deployments or plans you are aware of?

Srini


From: Christopher Price [mailto:christopher.price at est.tech]
Sent: Wednesday, February 13, 2019 6:20 AM
To: bharath thiruveedula <bharath_ves at hotmail.com>; edge-computing at lists.openstack.org
Subject: Re: [Edge-computing] Edge Node locator

Sorry I was “auto-corrected”:  CBN (content based networking), not CDN.

From: Christopher Price <christopher.price at est.tech<mailto:christopher.price at est.tech>>
Date: Wednesday, 13 February 2019 at 15:17
To: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>, "edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>" <edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>>
Subject: Re: [Edge-computing] Edge Node locator

Bharath,

A left field thought which I find quite interesting is to run your application on a CDN overlay.
This can be constrained to the application layer and run over IP which reduces the need to try and weld this type of function to the infrastructure layer or expose too much service awareness through an extended mesh.  (security security)  Moves the problem to a defined space and keeps your infrastructure and app workloads simple and focused on what they should be doing.

Millions of ways to skin that poor little cat though, imagination Vs cost…

/ Chris

From: bharath thiruveedula <bharath_ves at hotmail.com<mailto:bharath_ves at hotmail.com>>
Date: Tuesday, 12 February 2019 at 18:53
To: "edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>" <edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/1a96aa70/attachment.html>

From srinivasa.r.addepalli at intel.com  Wed Feb 13 15:06:27 2019
From: srinivasa.r.addepalli at intel.com (Addepalli, Srinivasa R)
Date: Wed, 13 Feb 2019 15:06:27 +0000
Subject: [Edge-computing] Edge Node locator
In-Reply-To: <BMXPR01MB2278D30C92EF3725D12FB6AA94660@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
References: <BMXPR01MB22785C483DA0D49A44E9A14594650@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
 <CF008BD37B7B9A4BA7FD12ECEC3C98016B0268B9@ORSMSX103.amr.corp.intel.com>,
 <DCE0B439-F523-4B5E-926D-61A233211C75@redhat.com>
 <BMXPR01MB2278D30C92EF3725D12FB6AA94660@BMXPR01MB2278.INDPRD01.PROD.OUTLOOK.COM>
Message-ID: <CF008BD37B7B9A4BA7FD12ECEC3C98016B028C73@ORSMSX103.amr.corp.intel.com>

Current solutions being discussed in open source projects are taking the approach of least disruption.  That is, no changes to end point operating systems and current networking technologies (IP Centric).  That is, why, there is middle broker (Application Function in the picture I shared).  End point application requests the broker, broker with the help of orchestrators places the endpoint application specific workloads if needed in the right edge, gets the reachability information of the workload in the edge and responds back to end point application with reachability information.   Since end-point application, broker, workloads are either  from the same vendor or from vendors with a business relationship (For example, gaming application needing to have access to gaming server), they have their own communication protocols for communication.

For generic application (such as browsers), I agree that there needs to be something else - such as UPF in case of 5G that can redirect the end application traffic to right workload in the edge or something else like NDN.

On ONAP:

-        ONAP OOF project has policy rules for cost, distance, HPA in R3 itself. They are used to determine the best possible edge location to place application specific workloads.  F-GPS feature that is being added in R4 is for taking care of affinity and anti-affinity rules, which may add to accurate placement.

Thanks
Srini

From: bharath thiruveedula [mailto:bharath_ves at hotmail.com]
Sent: Wednesday, February 13, 2019 6:08 AM
To: Azhar Sayeed <asayeed at redhat.com>; Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com>
Cc: edge-computing at lists.openstack.org; ramkik at vmware.com
Subject: Re: [Edge-computing] Edge Node locator

Hi Srini/Sayeed,

Thanks a lot for the interesting points

If not  UE directly detecting the edge node, I was looking for the solution where the service request from UE will be redirected to the edge.


  1.  Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those  based on certain criteria.
  2.  Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.
I think these will hold good, if edge is installed/deployed and configured to communicate  to Edge Orchestrator. We can also provision the edge from the Orchestrator itself. In that case, we don't need advertise and discover the Edge node and it will be part of the inventory. But I think the former case is the better option.


  1.  A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the "closest" edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
  2.  A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone" (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

For this process, we need changes in the UE I believe. Is there any documentation for more details?

Srini, I was looking for similar solutions in ONAP. The work you mentioned in OOF will be part of Dublin? Are you referring to FGPS?

Best Regards
Bharath T

________________________________
From: Azhar Sayeed <asayeed at redhat.com<mailto:asayeed at redhat.com>>
Sent: Wednesday, February 13, 2019 12:36 AM
To: Addepalli, Srinivasa R
Cc: bharath thiruveedula; edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>; ramkik at vmware.com<mailto:ramkik at vmware.com>
Subject: Re: [Edge-computing] Edge Node locator

The way I have been thinking about this is the following..

You will have to separate infra functions such as the Edge Nodes from service functions such as applications that run on edge nodes and provide a service.

>From an Infra management point of view..


  1.  Pro-active Advertisement by the Edge Node and its capacity so that Orchestration engines can place those work loads based on certain criteria.
  2.  Node Discovery Process - Explicit discovery of available capacity from the DC/operations center.

Both of them can be achieved via a service bus/message bus architecture that uses pub/sub concepts where you publish information and subscribe to topics. Edge nodes need to participate in that service bus model. It makes the discovery process and provisioning of applications to Edge nodes a lot easier and doesn't require yet another discovery protocol... This is the work best suited for Orchestrators - So perhaps ONAP

>From a UE point of view - A UE should not care where an edge node is located. UE does a normal attach process to the network. The network maps the available services and shows them to UE via a catalog or another mechanism. The UE decides to either attach to that service or sign up for push notifications. So


  1.  A UE can then do an explicit attach to a service that is being advertised as part of the catalog. The Edge node can service that request. If there are multiple Edge nodes that can serve the request then based on the infra policy the "closest" edge node can service it.. that would be an infra/orchestration rule as to which Edge node serves that request - the policy defined in the orchestration engine can take into account things such as latency, capacity etc etc.
  2.  A UE can sign up for push notifications of services - As the UE enters the proximity of the "Edge zone" (can be defined as a service area for that Edge compute/nodes) the local node can take action to provide the service based on subscription models

Bottom line, A UE should not care where the Edge node is.

Regards,
Azhar

On Feb 12, 2019, at 1:10 PM, Addepalli, Srinivasa R <srinivasa.r.addepalli at intel.com<mailto:srinivasa.r.addepalli at intel.com>> wrote:

Hi Bharath,

JFYI
This work is going on in ONAP (OOF project and Edge automation project).

Other possible projects are  MobileEdgeX and LFE EVE projects (These two are not yet open sourced).

ONAP OOF project intention is to figure out the best edge location for a given workload based on following criteria:

-        Latency/Distance with respect to UEs.
-        Availability of right hardware capabilities required for the workload.
-        Cost
-        Affinity and Anti-affinity rules
-        Etc...

Once ONAP OOF chooses the edge location to  place the workload, then ONAP (via Multi-Cloud project) will place the workload by calling Openstack API or K8S API etc...

You may want to check out ONAP project.

Thanks
Srini




Thanks
Srini


From: bharath thiruveedula [mailto:bharath_ves at hotmail.com]
Sent: Tuesday, February 12, 2019 9:52 AM
To: edge-computing at lists.openstack.org<mailto:edge-computing at lists.openstack.org>
Subject: [Edge-computing] Edge Node locator

Hi ,

Disclaimer: I am not sure the question following is in the scope of this forum.

I am trying to understand the Edge Architecture from UE point of view. How an app in UE can detect what the nearest edge it has to contact and how does it know whether the app is provisioned in that edge or not?

Can you please share some pointers from this perspective?

Best Regards
Bharath T
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org<mailto:Edge-computing at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190213/26d5592e/attachment-0001.html>

From ildiko at openstack.org  Thu Feb 14 09:18:49 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Thu, 14 Feb 2019 10:18:49 +0100
Subject: [Edge-computing] MEC presentation slides and recording
Message-ID: <E8DE9CF7-AF7B-401A-A34A-C0A9322EAE18@openstack.org>

Hi,

Attached please find the ETSI MEC slides from this week’s weekly call.

You can check out the recording from the meeting here: https://zoom.us/recording/share/dnRWoPhLvMJz1QUvsjMV7COhydppAwtc3J8w2q08VqmwIumekTziMw The recording starts at 00:16.

I also added a section to the wiki on ETSI MEC with a few pointers: https://wiki.openstack.org/wiki/Edge_Computing_Group#ETSI_MEC

Thanks,
Ildikó

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2019_02_12__ETSI_MEC_Overview_at_OSF - v2.pdf
Type: application/pdf
Size: 2919490 bytes
Desc: not available
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190214/0c90a8c9/attachment-0001.pdf>
-------------- next part --------------




From gergely.csatari at nokia.com  Tue Feb 19 11:20:56 2019
From: gergely.csatari at nokia.com (Csatari, Gergely (Nokia - HU/Budapest))
Date: Tue, 19 Feb 2019 11:20:56 +0000
Subject: [Edge-computing] LF Edge Glossary PR-s
Message-ID: <AM6PR07MB616707BC285B268D7367B4F1E17C0@AM6PR07MB6167.eurprd07.prod.outlook.com>

Hi,

Here are the pr-s what I'm proposing to the edge glossary:
https://github.com/lf-edge/glossary/pull/50
https://github.com/lf-edge/glossary/pull/49
https://github.com/lf-edge/glossary/pull/48
https://github.com/lf-edge/glossary/pull/47
https://github.com/lf-edge/glossary/pull/46
[https://avatars3.githubusercontent.com/u/11458775?s=400&v=4]<https://github.com/lf-edge/glossary/pull/46>

Adding a remark about cloud hw to Central Office by CsatariGergely · Pull Request #46 · lf-edge/glossary<https://github.com/lf-edge/glossary/pull/46>
New, cloud optimized hardware designs do not suffer from the constraints of a CEntral Office. It is important to add this to the Central Office section as this hardware design makes possible to reu...
github.com



Please check and comment them.

Br,
Gerg0
[https://avatars3.githubusercontent.com/u/11458775?s=400&v=4]<https://github.com/lf-edge/glossary/pull/47>

Correction to Cloud Computing by CsatariGergely · Pull Request #47 · lf-edge/glossary<https://github.com/lf-edge/glossary/pull/47>
The definition of Cloud Computing missed the networking resources and had network servers instead. Signed-off-by: Gergely Csatari gergely.csatari at nokia.com
github.com



[https://avatars3.githubusercontent.com/u/11458775?s=400&v=4]<https://github.com/lf-edge/glossary/pull/48>

Cloud ran by CsatariGergely · Pull Request #48 · lf-edge/glossary<https://github.com/lf-edge/glossary/pull/48>
Adding a references to DAS hub.
github.com


[https://avatars3.githubusercontent.com/u/11458775?s=400&v=4]<https://github.com/lf-edge/glossary/pull/49>

CPE definition clarifcation by CsatariGergely · Pull Request #49 · lf-edge/glossary<https://github.com/lf-edge/glossary/pull/49>
The CPE definition did not specify in which direction it is one hop away from cloud infra. Signed-off-by: Gergely Csatari gergely.csatari at nokia.com
github.com



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190219/fe810590/attachment.html>

From ildiko at openstack.org  Tue Feb 19 17:00:44 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Tue, 19 Feb 2019 18:00:44 +0100
Subject: [Edge-computing] Use cases weekly call in new time slot starts on
	February 25
Message-ID: <A20247D8-6E84-4A84-BE05-95874D0AF827@openstack.org>

Hi,

We are moving the use cases weekly calls back to a more popular slot.

The new time is every other Monday at 1pm PT / 2100 UTC: https://www.openstack.org/assets/edge/OSF-Edge-WG-Use-Cases-Weekly-Calls.ics

For dial-in info and agenda please see the wiki: https://wiki.openstack.org/wiki/Edge_Computing_Group#Use_cases

Please let me know if you have any questions.

Thanks and Best Regards,
Ildikó




From Arkady.Kanevsky at dell.com  Tue Feb 19 18:01:56 2019
From: Arkady.Kanevsky at dell.com (Arkady.Kanevsky at dell.com)
Date: Tue, 19 Feb 2019 18:01:56 +0000
Subject: [Edge-computing] Use cases weekly call in new time slot starts
	on	February 25
In-Reply-To: <A20247D8-6E84-4A84-BE05-95874D0AF827@openstack.org>
References: <A20247D8-6E84-4A84-BE05-95874D0AF827@openstack.org>
Message-ID: <fb84bffd713d442880ee5ae28dae0746@AUSX13MPS308.AMER.DELL.COM>

Is this replacement for Tu call (morning US time)?

-----Original Message-----
From: Ildiko Vancsa <ildiko at openstack.org> 
Sent: Tuesday, February 19, 2019 11:01 AM
To: edge-computing
Subject: [Edge-computing] Use cases weekly call in new time slot starts on February 25


[EXTERNAL EMAIL] 

Hi,

We are moving the use cases weekly calls back to a more popular slot.

The new time is every other Monday at 1pm PT / 2100 UTC: https://www.openstack.org/assets/edge/OSF-Edge-WG-Use-Cases-Weekly-Calls.ics

For dial-in info and agenda please see the wiki: https://wiki.openstack.org/wiki/Edge_Computing_Group#Use_cases

Please let me know if you have any questions.

Thanks and Best Regards,
Ildikó



_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/edge-computing

From beth.cohen at verizon.com  Tue Feb 19 19:43:10 2019
From: beth.cohen at verizon.com (beth.cohen at verizon.com)
Date: Tue, 19 Feb 2019 19:43:10 +0000
Subject: [Edge-computing] [E] Re: Use cases weekly call in new time slot
 starts	on	February 25
In-Reply-To: <fb84bffd713d442880ee5ae28dae0746@AUSX13MPS308.AMER.DELL.COM>
References: <A20247D8-6E84-4A84-BE05-95874D0AF827@openstack.org>
 <fb84bffd713d442880ee5ae28dae0746@AUSX13MPS308.AMER.DELL.COM>
Message-ID: <03b6ba351bfd4f09951af7119d3dd892@scwexch25apd.uswin.ad.vzwcorp.com>

No.   This is the continuation of the Edge Use case meeting going back to the old time.




Beth Cohen
DMTS-  NFV/SDN Network Product Strategy
Verizon Technology

O +1-781-466-2055 | M +1-781-434-8553
beth.cohen at verizon.com

    


-----Original Message-----
From: Arkady.Kanevsky at dell.com [mailto:Arkady.Kanevsky at dell.com] 
Sent: Tuesday, February 19, 2019 1:02 PM
To: ildiko at openstack.org; edge-computing at lists.openstack.org
Subject: [E] Re: [Edge-computing] Use cases weekly call in new time slot starts on February 25

Is this replacement for Tu call (morning US time)?

-----Original Message-----
From: Ildiko Vancsa <ildiko at openstack.org> 
Sent: Tuesday, February 19, 2019 11:01 AM
To: edge-computing
Subject: [Edge-computing] Use cases weekly call in new time slot starts on February 25


[EXTERNAL EMAIL] 

Hi,

We are moving the use cases weekly calls back to a more popular slot.

The new time is every other Monday at 1pm PT / 2100 UTC: https://urldefense.proofpoint.com/v2/url?u=https-3A__www.openstack.org_assets_edge_OSF-2DEdge-2DWG-2DUse-2DCases-2DWeekly-2DCalls.ics&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=K1lvHRH7aF_9Rcr3ufyM0mIapj6RSK408Rj3rytWuAE&e=

For dial-in info and agenda please see the wiki: https://urldefense.proofpoint.com/v2/url?u=https-3A__wiki.openstack.org_wiki_Edge-5FComputing-5FGroup-23Use-5Fcases&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=yNKcvJrEaKSMPc_T4OKbVkawY0sR8J8puR2_u9FxDyQ&e=

Please let me know if you have any questions.

Thanks and Best Regards,
Ildikó



_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org
https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.openstack.org_cgi-2Dbin_mailman_listinfo_edge-2Dcomputing&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=FatjBbB0UJ74ob_BYxS9OdNNzN3ws_OZQOpsYCNosXg&e=
_______________________________________________
Edge-computing mailing list
Edge-computing at lists.openstack.org
https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.openstack.org_cgi-2Dbin_mailman_listinfo_edge-2Dcomputing&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=FatjBbB0UJ74ob_BYxS9OdNNzN3ws_OZQOpsYCNosXg&e=

From ildiko at openstack.org  Tue Feb 19 20:37:02 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Tue, 19 Feb 2019 21:37:02 +0100
Subject: [Edge-computing] Open Infrastructure Summit Forum and PTG
	preparations
Message-ID: <DF8D06A9-2D81-4030-AF35-2C0A496F262C@openstack.org>

Hi,

The Open Infrastructure Summit & PTG in Denver is approaching quickly so it is time to start preparing for the Forum and think of topics for the PTG as well.

I created an etherpad to collect Forum session ideas: https://etherpad.openstack.org/p/edge-wg-forum-preparation-denver-2019

The Forum session proposal period opens on February 22 and closes on March 8. For the detailed process and deadlines please see the following wiki: https://wiki.openstack.org/wiki/Forum

Please add you ideas to the etherpad above and add +1’s for other ones you like and -1’s that you don’t think being a good fit for the Forum. We can revisit and finalize planning next week and the session moderators can go ahead and submit the sessions through the CFP tool.


We’ve decided to request for a half day slot at the PTG for Edge WG specific topics and in the remaining time sit with project teams for discussing relevant items. To make sure we spend the time available to discuss important topics I created another etherpad to collect ideas and start planning: https://etherpad.openstack.org/p/edge-wg-ptg-preparation-denver-2019


Please check the etherpads and add your ideas and comments there. Let me know if you have any questions.

Thanks and Best Regards,
Ildikó




From ildiko at openstack.org  Tue Feb 19 20:58:44 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Tue, 19 Feb 2019 21:58:44 +0100
Subject: [Edge-computing] [E] Use cases weekly call in new time slot
 starts	on	February 25
In-Reply-To: <03b6ba351bfd4f09951af7119d3dd892@scwexch25apd.uswin.ad.vzwcorp.com>
References: <A20247D8-6E84-4A84-BE05-95874D0AF827@openstack.org>
 <fb84bffd713d442880ee5ae28dae0746@AUSX13MPS308.AMER.DELL.COM>
 <03b6ba351bfd4f09951af7119d3dd892@scwexch25apd.uswin.ad.vzwcorp.com>
Message-ID: <B4C66C41-5B14-466A-A859-D7C383EDFB99@openstack.org>

Hi Arkady,

As Beth said the weekly working group calls on Tuesdays (and sometimes Thursdays :) remain the same.

We’ve been having the Use cases sub-call with the purpose of collecting more details on use cases and have a more focused discussion around this topic than what the weekly WG calls allow.

We switched over to alternating time slots which didn’t work out therefore we are moving back the calls to the old time slot with a bi-weekly cadence. We will report on progress and activities on the weekly WG calls.

I added defining goals for the sub-group as one of the agenda items to make sure we have a clear direction to work towards and then we’ll revisit the current stage and decide on next steps.

Please let me know if you have further questions.

Thanks,
Ildikó


> On 2019. Feb 19., at 20:43, beth.cohen at verizon.com wrote:
> 
> No.   This is the continuation of the Edge Use case meeting going back to the old time.
> 
> 
> 
> 
> Beth Cohen
> DMTS-  NFV/SDN Network Product Strategy
> Verizon Technology
> 
> O +1-781-466-2055 | M +1-781-434-8553
> beth.cohen at verizon.com
> 
>     
> 
> 
> -----Original Message-----
> From: Arkady.Kanevsky at dell.com [mailto:Arkady.Kanevsky at dell.com] 
> Sent: Tuesday, February 19, 2019 1:02 PM
> To: ildiko at openstack.org; edge-computing at lists.openstack.org
> Subject: [E] Re: [Edge-computing] Use cases weekly call in new time slot starts on February 25
> 
> Is this replacement for Tu call (morning US time)?
> 
> -----Original Message-----
> From: Ildiko Vancsa <ildiko at openstack.org> 
> Sent: Tuesday, February 19, 2019 11:01 AM
> To: edge-computing
> Subject: [Edge-computing] Use cases weekly call in new time slot starts on February 25
> 
> 
> [EXTERNAL EMAIL] 
> 
> Hi,
> 
> We are moving the use cases weekly calls back to a more popular slot.
> 
> The new time is every other Monday at 1pm PT / 2100 UTC: https://urldefense.proofpoint.com/v2/url?u=https-3A__www.openstack.org_assets_edge_OSF-2DEdge-2DWG-2DUse-2DCases-2DWeekly-2DCalls.ics&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=K1lvHRH7aF_9Rcr3ufyM0mIapj6RSK408Rj3rytWuAE&e=
> 
> For dial-in info and agenda please see the wiki: https://urldefense.proofpoint.com/v2/url?u=https-3A__wiki.openstack.org_wiki_Edge-5FComputing-5FGroup-23Use-5Fcases&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=yNKcvJrEaKSMPc_T4OKbVkawY0sR8J8puR2_u9FxDyQ&e=
> 
> Please let me know if you have any questions.
> 
> Thanks and Best Regards,
> Ildikó
> 
> 
> 
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.openstack.org_cgi-2Dbin_mailman_listinfo_edge-2Dcomputing&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=FatjBbB0UJ74ob_BYxS9OdNNzN3ws_OZQOpsYCNosXg&e=
> _______________________________________________
> Edge-computing mailing list
> Edge-computing at lists.openstack.org
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.openstack.org_cgi-2Dbin_mailman_listinfo_edge-2Dcomputing&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=zSPYRgoBij7eDdloAJpktY4pyBQRGnvgrz_9Wy3fyA4&m=Ul82n1Ta1hcac_8-mldHbFDNqutSwQp1cmVor9LL-WM&s=FatjBbB0UJ74ob_BYxS9OdNNzN3ws_OZQOpsYCNosXg&e=



From ildiko at openstack.org  Mon Feb 25 20:23:36 2019
From: ildiko at openstack.org (Ildiko Vancsa)
Date: Mon, 25 Feb 2019 21:23:36 +0100
Subject: [Edge-computing] Uses cases weekly call in a half an hour
Message-ID: <088475D3-73D9-413B-8FC0-EA18683C815E@openstack.org>

Hi,

It is a friendly reminder that the next Uses cases sub-group call is at 1pm PT / 2100 UTC.

Call details and agenda are here: https://wiki.openstack.org/wiki/Edge_Computing_Group#Use_cases

Please let me know if you have any questions.

Thanks,
Ildikó




From aheczko at mirantis.com  Tue Feb 12 22:02:59 2019
From: aheczko at mirantis.com (aheczko at mirantis.com)
Date: Tue, 12 Feb 2019 22:02:59 -0000
Subject: [Edge-computing] [keystone] x509 authentication
In-Reply-To: <de099fdf-9cf6-b799-0404-61af51c0a1eb@gmail.com>
References: <CAE6oFcEWjZ1i0f=gfTZyas6o=xoaH05KjRPFWiZyr1=FygwzvA@mail.gmail.com>
 <de099fdf-9cf6-b799-0404-61af51c0a1eb@gmail.com>
Message-ID: <CAJciqMzFx1UFFf0vkNW3TEVaqi2+++Voaf_DULMhTN8rUYXbHQ@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Hi Lance, I'd be glad to help out with the docs update.
-----BEGIN PGP SIGNATURE-----
Version: FlowCrypt 6.6.1 Gmail Encryption
Comment: Seamlessly send and receive encrypted email
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=gitY
-----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0xA86B279C7109F5D6.asc
Type: application/pgp-keys
Size: 3177 bytes
Desc: not available
URL: <http://lists.openstack.org/pipermail/edge-computing/attachments/20190212/88bbc664/attachment.key>