[Openinfralabs] What operational data and data streams should be gathered?

Marcel Hild mhild at redhat.com
Thu Apr 16 10:52:28 UTC 2020


I would start with monitoring data, i.e. all prometheus metrics being
collected.
But those in itself are not really useful without a picture of the
infrastructure and access to the infrastructure. And you might need access
to ticket systems, since you want to correlate the metrics with incidents.

My point is, we need to open up _all_ aspects of operations to make it
actually useful, otherwise it'll be just a pile of data.

I was hoping that the newly build out NERC could be the blueprint for this.

On Thu, Apr 16, 2020 at 3:15 AM Michael Daitzman <msd at bu.edu> wrote:

>
> TL;DR:
>
> As we think about making operational data from the Mass Open Cloud
> <https://massopen.cloud> available to the open source community, we need
> to define what data and data streams will be useful and gathered.
>
> This email is meant to begin a discussion and define that.
>
> TS;WM:
>
> Some context - the Mass Open Cloud (MOC) is a research cloud formed as a
> public-private partnership between 5 Research Universities, a group of
> partner companies, and the state of Massachusetts.
>
> The MOC has always had a goal of sharing operational data about the MOC to
> enable research on cloud computing technologies.
>
> As teams working on OI Labs projects began discussing operational data,
> one question was, “can the MOC share that data to the open source
> community?”
>
> To a great extent, this becomes part of a “Research Compliance” question
> and may be affected by various grants which have funded the MOC (good news
> - in almost every grant the PI’s have made clear that data sharing is a
> goal).
>
> As we looked into making such information available, I had a brief
> discussion with the director of BU’s Research Computing group who helped
> frame how we should approach this question. The first steps are:
>
>    1.
>
>    Defining what data and data streams will be useful and gathered is the
>    first step
>    2.
>
>    Based on that we may identify which populations may be affected
>    3.
>
>    Decide on the next steps based on the answers to item 1 and 2 above
>
> _______________________________________________
> Openinfralabs mailing list
> Openinfralabs at lists.opendev.org
> http://lists.opendev.org/cgi-bin/mailman/listinfo/openinfralabs
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opendev.org/pipermail/openinfralabs/attachments/20200416/b015a111/attachment-0001.html>


More information about the Openinfralabs mailing list