# RLLib Parallelization Guide

**URL:** https://discuss.ray.io/t/rllib-parallelization-guide/15381
**Category:** Configure Algorithm, Training, Evaluation, Scaling
**Created:** [July 29, 2024, 9:48pm UTC](https://discuss.ray.io/t/rllib-parallelization-guide/15381 "2024-07-29T21:48:16Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![PhilippWillms](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/philippwillms/32/3885_2.png) [@PhilippWillms](https://discuss.ray.io/u/PhilippWillms)
#### Post date: [July 29, 2024, 9:48pm UTC](https://discuss.ray.io/t/rllib-parallelization-guide/15381/1 "2024-07-29T21:48:16Z")

</div>

Dear community,  
due to lots of work on my custom Gym environment, I can run much more iterations with bigger training batch sizes and SGD mini-batch sizes than before. Still, it is quite time-consuming. Would there be any parallelization guide available for RllIb (old stack) ? It could contain topic like how many rollout workers make sense in relation to the iteration count / batch size, whatever.