-
Currently AURA creates a worker pool based on the number of cores in the system. This was always meant to be a temporary approach.
The size of the worker pool should be configurable from the comman…
-
As far as I understand, the current way to use parallel iterators with a dedicated thread pool is to use `ThreadPool::install` and use the iterator inside of the provided callback, or use `ThreadPool:…
nical updated
6 years ago
-
### Search before asking
- [X] I searched the [issues](https://github.com/ray-project/kuberay/issues) and found no similar issues.
### KubeRay Component
ray-operator
### What happened + …
-
Something strange showed up in [ThreadSanitizer](https://ocaml.org/manual/latest/tsan.html) logs: in `Lwt_unix`, more precisely in `lwt_unix_start_job`, the `thread_waiting_count` variable is written …
-
Hi,
i want to edit this and i want to enable multiple pool for example some hours work for other pool and worker or split hashrate between two pool..
-
### What did you do?
1. master-0 failover to master-1
2. previous worker commit suicide
3. start a new worker but the worker stopped quickly.
4. repeat step 3 forever
### What did you expect …
-
### What happened + What you expected to happen
The below code fails with
```
(ReadParquet->MapBatches(preprocess) pid=14184) warnings.warn( [repeated 11x across cluster] …
-
When using a model endpoint for `precheck` and the `sdg-svc` for `generate`, we should be able to have multiple concurrent requests to these endpoints to help scale this out. Some thoughts on how to …
-
Worker failed to lock a job level=error worker-pool-id=959ce5 worker-id=959ce5/worker-0 error=could not lock a job (rollback result: ): ERROR: relation "gue_jobs" does not exist (SQLSTATE 42P01)
kulak updated
2 years ago
-
webui's wsproxy currently determines proxy worker's listening port by [choosing 100 random integers](https://github.com/lablup/backend.ai-webui/blob/main/src/wsproxy/manager.js#L56). We need to rewrit…