# Required resources should be shared between train and eval workers **URL:** \ **Category:** RLlib\ **Created:** [March 29, 2021, 4:07pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457 "2021-03-29T16:07:02Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![vakker00](https://avatars.discourse-cdn.com/v4/letter/v/b5e925/32.png) [@vakker00](https://discuss.ray.io/u/vakker00)\ **Post date:** [March 29, 2021, 4:07pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/1 "2021-03-29T16:07:02Z")
If I set `"num_workers": 10` and `"evaluation_num_workers": 10` then the required CPUs to start the training is going to be 20, but the 10-10 environment workers are used in sequence, i.e. the average utilisation is going to be 50% on a 20 CPU machine. Is there a way around this? I set `"num_cpus_per_worker": 0.5` with 20 train and eval workers which runs, but this it not ideal when there are other trials scheduled. ---
**Author:** ![yic](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/yic/32/437_2.png) [@yic](https://discuss.ray.io/u/yic)\ **Post date:** [March 29, 2021, 10:22pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/2 "2021-03-29T22:22:02Z")
> [@vakker00](#): > > If I set `"num_workers": 10` and `"evaluation_num_workers": 10` then the required CPUs to start the training is going to be 20, but the 10-10 environment workers are used in sequence, i.e. the average utilisation is going to be 50% on a 20 CPU machine. @sven1977 could you take a look at this? ---
**Author:** ![sven1977](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/sven1977/32/53_2.png) [@sven1977](https://discuss.ray.io/u/sven1977)\ **Post date:** [March 31, 2021, 1:36pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/3 "2021-03-31T13:36:41Z")
Hey @vakker00 , great question. Just confirmed this myself: Trainer.train() and Trainer.evaluate() execute in sequence, never in parallel, even though you can - of course - parallelize the needed evaluation rollouts by specifying `evaluation_num_workers > 1`. So yes, we will block all these CPUs, even though, we don’t need them all at the same time. I wonder whether it would make sense to run evaluation in a separate thread (after having synched all weights to avoid race conditions). I’ll play around with this idea and get back to you. … ---
**Author:** ![sven1977](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/sven1977/32/53_2.png) [@sven1977](https://discuss.ray.io/u/sven1977)\ **Post date:** [March 31, 2021, 1:37pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/4 "2021-03-31T13:37:29Z")
The other option would be to leave it sequential and tell tune to not reserve CPUs for evaluation (or reserve the max between `num_workers` and `evaluation_num_workers`). ---
**Author:** ![sven1977](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/sven1977/32/53_2.png) [@sven1977](https://discuss.ray.io/u/sven1977)\ **Post date:** [March 31, 2021, 2:25pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/5 "2021-03-31T14:25:07Z")
Got it to run in parallel now (optional via a new config flag). I’ll PR. If not using parallelism, I’ll make the Trainer return reduced CPU resource requirements (b/c we never use evaluation workers and rollout workers at the same time). PR: [[RLlib] Support parallelizing evaluation and training (optional). by sven1977 · Pull Request #15040 · ray-project/ray · GitHub](https://github.com/ray-project/ray/pull/15040) The above PR will allow you to run evaluation and training in parallel. Btw, we won’t be able to CPU-share between eval + train workers due to the restriction of having one actor per CPU. Tune would hang otherwise so the option I described above (for non-parallelized eval\_training) won’t work. In other words, you will still need 20 CPUs for your setup. ---
**Author:** ![vakker00](https://avatars.discourse-cdn.com/v4/letter/v/b5e925/32.png) [@vakker00](https://discuss.ray.io/u/vakker00)\ **Post date:** [March 31, 2021, 3:38pm UTC](https://discuss.ray.io/t/required-resources-should-be-shared-between-train-and-eval-workers/1457/6 "2021-03-31T15:38:55Z")
Thanks @sven1977 for looking into this. So just to confirm: using `"num_cpus_per_worker": 0.5` won’t work because of the one actor/CPU restriction? I did manage to run a training with 10 workers and 0.5 CPU/worker and the requested resources is 21.5/24 CPU (I have 24 CPUs, not 20), which is (if I understand it correctly) 0.5\*2\*10+0.5+1, so two times 10 for the rollout and eval weighted 0.5, 0.5 for the model and 1 for the main process? If I remove the `"num_cpus_per_worker": 0.5` then it doesn’t run. So with the new PR the evaluation can run with (let’s say) 2 CPUs while the new training sampling can run with 18 (requiring 20 CPUs), right? Would this cause issues in case the eval doesn’t finish before it’s time for a new weight sync?