# MPI framework for parallelized warmups

**URL:** <https://discourse.mc-stan.org/t/mpi-framework-for-parallelized-warmups/11671>\
**Category:** Algorithms\
**Tags:** mcmc\
**Created:** [October 30, 2019, 10:58pm UTC](https://discourse.mc-stan.org/t/mpi-framework-for-parallelized-warmups/11671 "2019-10-30T22:58:51Z")\
**Posts on this page:** 1\
**Showing post:** 9

<div class="post-metadata">

**Author:** ![yizhang](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/yizhang/32/15673_2.png) [@yizhang](https://discourse.mc-stan.org/u/yizhang)\
**Post date:** [November 1, 2019, 3:04pm UTC](https://discourse.mc-stan.org/t/mpi-framework-for-parallelized-warmups/11671/9 "2019-11-01T15:04:23Z")

</div>

> [@wds15](#):
>
> MPI is not available on Windows for stan at the moment since I never managed to get boost mpi working on that platform

As I pointed out at early stage of MPI, [we don’t need boost](https://discourse.mc-stan.org/t/mpi-design-discussion/1103/263). All my MPI implementation doesn’t require boost MPI and works on MS MPI.

> [@wds15](#):
>
> Managing resources for parallelisation is something I would ideal just leave to the TBB.

How does TBB manage distributed parallel resource?

> [@wds15](#):
>
> We should probably start a new discourse thread for that.

Yes. I’ve said this [earlier](https://discourse.mc-stan.org/t/evaluating-parallelization-performance/10962/3), we’re not doing things in a coordinated way.

---

_[View the full topic](https://discourse.mc-stan.org/t/mpi-framework-for-parallelized-warmups/11671)._