Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create a local perf test runner #25872

Open
Tracked by #25870
praveen-influx opened this issue Jan 20, 2025 · 0 comments
Open
Tracked by #25870

Create a local perf test runner #25872

praveen-influx opened this issue Jan 20, 2025 · 0 comments
Labels

Comments

@praveen-influx
Copy link
Contributor

praveen-influx commented Jan 20, 2025

Problem statement

Need a way to run perf test locally (e.g dev laptop) without overwhelming the host system. This would allow devs to tune the performance by using tools available in the host OS. In Linux it can be perf, heaptrack etc (depending on what needs to be captured). The issues especially related to OOMs or CPU too high should be traceable to the actual code path (by attaching debug symbols in release profile).

Proposed solution (TBD)

Options available

  • Use docker-compose or just plain docker to run influxdb in one of the containers and the load generator in another container with hard limits for memory and cpu setup at the docker level for both of them.

  • Run influxdb3 and load generator on the host system, by creating cgroups to be able to control the system resources used by each process. influxdb3 for example can be configured to take up X number of CPU quotas/slices and use Y as the memory allowance.

Using docker-compose for example allows it to be used in macos as well. However, it would still require Linux tools to run any perf testing within the container (as explained here). Natively running on Linux (or macos) for example should be helpful as it is easier to run any platform specific tools to diagnose perf issues.

Alternatives considered

TBD

Additional context

N/A

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant