[External] : Re: Project Loom VirtualThreads hang

[Pedro Lamarão]

Em qui., 5 de jan. de 2023 às 16:04, Sam Pullara <spullara at gmail.com>
escreveu:

> So the place where this comes up is when you do a short computation but
> use tons of threads to do it. Like, for example, I have a brute force
> embedding search library that uses parallel streams to do the query. If
> that takes a few seconds (or even hundreds of ms) I still want other
> threads doing more mundane, fast work to get CPU time rather than every
> other computation on the system being blocked.
>

If this system processes more than one request at a time, you will not want
any single request to occupy 100% processor time anyway. If you do have
spare processor time, those fast tasks will get done. If you do not, your
system has crossed the line to instability, and all tasks, not only the
fast ones, will receive a penalty. In cases like these, time sharing, like
disk swapping, is a last resort to keep the system running.

-- 
Pedro Lamarão
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/loom-dev/attachments/20230105/0161ff5c/attachment-0001.htm>