Request - Equinix bare metal to use for OpenTelemetry benchmark performance tests

[cartersocha]

First and Last Name

Carter Socha Ted Young

Email

carter.socha@servicenow.com theodore.young@servicenow.com

Company/Organization

ServiceNow / Lightstep

Job Title

OpenTelemetry PM Director of DevRel / OpenTelemetry Co-Founder / OpenTelemetry Governance Committee

Project Title (i.e., a summary of what do you want to do, not what is the name of the open source project you're working with)

OpenTelemetry Component Performance Benchmarking

Briefly describe the project (i.e., what is the detail of what you're planning to do with these servers?)

OpenTelemetry has seen tremendous growth and adoption over the years but as the project matures there are consistent questions raised about the performance of the project components. Our goal is to help maintainers build efficient components by highlighting current performance and preventing performance regressions while giving customers meaningful data about the potential performance impact when they use OpenTelemetry in their services.

To provide a consistent benchmark we need standardized machines and underlying runner performance. GitHub standard runners are too variable to be useful. The eventual goal of this project is to define OpenTelemetry performance standards and roll out performance testing to all OpenTelemetry components.

Is the code that you’re going to run 100% open source? If so, what is the URL or URLs where it is located? What is your association with that project?

100% open source https://github.com/open-telemetry/opentelemetry-collector https://github.com/open-telemetry/opentelemetry-js https://github.com/open-telemetry/opentelemetry-java https://github.com/open-telemetry/opentelemetry-python

Carter is a maintainer of several community projects and Ted is a co-founder / governance committee member

What kind of machines and how many do you expect to use (see: https://metal.equinix.com/product/servers/)?

2 machines. Count may grow as we experiment with supporting a wider array of tests and components in OpenTelemetry. https://deploy.equinix.com/product/servers/m3-small/ x2

What operating system and networking are you planning to use?

Linux

Any other relevant details we should know about?

[cartersocha]

@tedsuo

[jpkrohling]

Not sure this is needed, but I support this request.

Edit: "this" refers to my support. The machines are definitely needed :-)

[caniszczyk]

+1

On Fri, Jul 28, 2023 at 3:24 AM Juraci Paixão Kröhling < @.***> wrote:

Not sure this is needed, but I support this request.

— Reply to this email directly, view it on GitHub https://github.com/cncf/cluster/issues/245#issuecomment-1654838579, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAPSIMPENVKDUDH4CPUMF3XSMIE3ANCNFSM6AAAAAA22RSCYM . You are receiving this because you were assigned.Message ID: @.***>

-- Cheers,

Chris Aniszczyk https://aniszczyk.org

[cartersocha]

@caniszczyk any additional information or follow up needed from me?

[jeefy]

Nope we'll get you hooked up soon! 🙂

On Tue, Aug 1, 2023, 12:41 Carter Socha @.***> wrote:

@caniszczyk https://github.com/caniszczyk any additional information or follow up needed from me?

— Reply to this email directly, view it on GitHub https://github.com/cncf/cluster/issues/245#issuecomment-1660801836, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKUO6DRSVFBIHEK2JHKTJDXTE5UTANCNFSM6AAAAAA22RSCYM . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[cartersocha]

Quick ping on this. Our testing pieces are ready besides stable compute

[caniszczyk]

+1 @jeefy @RobertKielty can you take care of this

[RobertKielty]

Invites sent out to both email addrs. Please get Carter and Ted to check their mail and accept the invites. Any issues reach out to me straight away.

[cartersocha]

@RobertKielty would it be possible to get access to add other contributors to the project or is the preferred method I share my credentials? We have a dozen or so separate SIGs that will use the compute

[RobertKielty]

@caniszczyk @jeefy what would be the preferred course of action here?

[vielmetti]

My preference from the Equinix side (and your preference should be too) is that individuals should have individual credentials - you'll want that to track responsible parties for use. Sharing a login always gets weird when people leave a project for whatever reason.

[vielmetti]

@cartersocha @tedsuo Happy to talk through any questions you have + get you spun up on any of the Equinix tooling that could help with automation. I'm also tracking https://github.com/open-telemetry/community/issues/1617 which looks like the upstream with some of the other contributors?

[cartersocha]

Yes that's correct on the issue. Sharing credentials is definitely not preferred. However, to create individual credentials for others I need the ability to invite project contributors. At the moment access is solely on my personal account and I can't share the love by sending out invites.

[vielmetti]

@cartersocha We have done this in the past by providing a list of individuals to invite (or to uninvite, if/when that's the case) to the CNCF project team who are empowered to send the invites out and grant access.

[jpkrohling]

@vielmetti, those people would then be me and @cartersocha. It looks like I'm already part of the project, as it seems to have been reused from a previous Outreachy internship. I have an old email address as my login data there, but I can still log in.

[vielmetti]

@jpkrohling @cartersocha from additional correspondence it sounds like this is all taken care of now, but if it's not please let us know; I'd like to wrap this issue up once everyone who needs access has it.

[cartersocha]

We can go ahead and close this. Thanks all for the support and help! We’ll follow up on slack if anything else is needed.