# Monadic concurrency vs algebraic effects

**URL:** https://discuss.ocaml.org/t/monadic-concurrency-vs-algebraic-effects/6395
**Category:** Ecosystem
**Tags:** multicore
**Created:** [September 11, 2020, 11:21am UTC](https://discuss.ocaml.org/t/monadic-concurrency-vs-algebraic-effects/6395 "2020-09-11T11:21:52Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![Cjen1](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ocaml.org/cjen1/32/1731_2.png) [@Cjen1](https://discuss.ocaml.org/u/Cjen1)
#### Post date: [September 11, 2020, 11:21am UTC](https://discuss.ocaml.org/t/monadic-concurrency-vs-algebraic-effects/6395/1 "2020-09-11T11:21:52Z")

</div>

I know that there is lots of great work being done on multicore with regards to performance benchmarking. I was wondering what the current state of performance of effects was (in comparison to Lwt)?

I found this blog post from 2011 [https://mirage.io/blog/delimcc-vs-lwt](https://mirage.io/blog/delimcc-vs-lwt) comparing delimcc (which I believe that algebraic effects was based on?) and lwt.

The outcome from that seemed to be inconclusive, though it appears that the direction multicore ocaml is going, is towards algebraic effects. Is there a more modern performance benchmark comparing them?

(My current use case is in the webserver space, if that changes the analysis)

---

<div class="post-metadata">

### Author: ![kayceesrk](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ocaml.org/kayceesrk/32/1936_2.png) [@kayceesrk](https://discuss.ocaml.org/u/kayceesrk)
#### Post date: [September 11, 2020, 12:55pm UTC](https://discuss.ocaml.org/t/monadic-concurrency-vs-algebraic-effects/6395/2 "2020-09-11T12:55:24Z")

</div>

See Fig.2 in [https://kcsrk.info/papers/system\_effects\_feb\_18.pdf](https://kcsrk.info/papers/system_effects_feb_18.pdf) where we compare Async vs Golang vs Effect Handlers on a webserver.
