# First call latency in distributed code

**URL:** https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917
**Category:** Performance
**Tags:** distributed, latency
**Created:** [October 18, 2022, 3:07pm UTC](https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917 "2022-10-18T15:07:32Z")
**Posts on this page:** 3
**Page:** 3

<div class="post-metadata">

### Author: ![dlakelan](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/dlakelan/32/8491_2.png) [@dlakelan](https://discourse.julialang.org/u/dlakelan)
#### Post date: [October 20, 2022, 4:11pm UTC](https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917/41 "2022-10-20T16:11:33Z")

</div>

> [@shce](#):
>
> v1.6 since it is the LTS currently.

The only people who should use LTS is basically people who have a running public service based on Julia that they want to keep running continuously without modifications for years, and they just want security updates and bugfixes. If that’s not you, use the latest release.

---

<div class="post-metadata">

### Author: ![jmair](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jmair/32/35117_2.png) [@jmair](https://discourse.julialang.org/u/jmair)
#### Post date: [October 20, 2022, 7:30pm UTC](https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917/42 "2022-10-20T19:30:37Z")

</div>

I agree with @jling and @dlakelan, `v1.8+` is the way to go.

---

<div class="post-metadata">

### Author: ![shce](https://avatars.discourse-cdn.com/v4/letter/s/d78d45/32.png) [@shce](https://discourse.julialang.org/u/shce)
#### Post date: [October 21, 2022, 11:52am UTC](https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917/43 "2022-10-21T11:52:38Z")

</div>

Hello,

It seems that the problem was a not properly adjusted tolerance, which prevented the optimization to stop. I thought it was something related to parallel computations because I was experiencing large timings and nothing was printed (even though I used `flush(stdout)` or even after cancelling the job). My apologies.

In any case, I have learnt that the code is compiled for each worker separately, and I have left the fear behind creating packages, sysimages and having two julia versions installed.

Thank you very much.

shce

[Previous page](https://discourse.julialang.org/t/first-call-latency-in-distributed-code/88917.md?page=2)