# RFC: Make an abstraction for custom task parallel schedulers (GSoC proposal)

**URL:** <https://discourse.julialang.org/t/rfc-make-an-abstraction-for-custom-task-parallel-schedulers-gsoc-proposal/33284>\
**Category:** Internals & Design\
**Tags:** multithreading\
**Created:** [January 13, 2020, 12:08am UTC](https://discourse.julialang.org/t/rfc-make-an-abstraction-for-custom-task-parallel-schedulers-gsoc-proposal/33284 "2020-01-13T00:08:47Z")\
**Posts on this page:** 1\
**Showing post:** 14

<div class="post-metadata">

**Author:** ![tkf](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/tkf/32/17635_2.png) [@tkf](https://discourse.julialang.org/u/tkf)\
**Post date:** [January 15, 2020, 5:08am UTC](https://discourse.julialang.org/t/rfc-make-an-abstraction-for-custom-task-parallel-schedulers-gsoc-proposal/33284/14 "2020-01-15T05:08:39Z")

</div>

Yeah, that’s related to the fancier solution I was thinking. It’d be nice to hook “splittable” PRNGs in the `@spawn` mechanism and make it task-local than thread-local. I mentioned it at [Brew a Parallel RNG? - #5 by tkf](https://discourse.julialang.org/t/brew-a-parallel-rng/30286/5) before (see also [Parallel Mersenne Twister - #15 by oschulz](https://discourse.julialang.org/t/parallel-mersenne-twister/27567/15)). Providing this via an effect system sounds like an interesting and possibly elegant solution.

---

_[View the full topic](https://discourse.julialang.org/t/rfc-make-an-abstraction-for-custom-task-parallel-schedulers-gsoc-proposal/33284)._