apache / datafusion-ballista

Apache DataFusion Ballista Distributed Query Engine

Home Page:https://datafusion.apache.org/ballista

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[LOG] Print long running task status.

Ted-Jiang opened this issue · comments

Is your feature request related to a problem or challenge? Please describe what you are trying to do.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
(This section helps Arrow developers understand the context and why for this feature, in addition to the what)
I use a atomicU64 as a counter to calculate all events cost time in while loop

https://github.com/Ted-Jiang/arrow-ballista/blob/0b6b28f8d2bdeca8f2d66e5453008dca11974dae/ballista/core/src/event_loop.rs#L78-L88

I run 5 mins load test got 2023-04-14T07:52:45.743867Z WARN tokio-runtime-worker ThreadId(16) ballista_scheduler::scheduler_server::query_stage_scheduler: [METRICS] total event cost 278 955 949 us

So the events processing is the hot path and bottleneck in scheduler. we need log the long running event status.

Describe the solution you'd like
Add a config scheduler_event_expected_processing_duration if event processing time more than this warn in log.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.