OpenVZ Forum


Home » Mailing lists » Devel » Re: [Lxc-devel] Q: Do systems using containers user more process ids?
Re: [Lxc-devel] Q: Do systems using containers user more process ids? [message #5125] Mon, 14 August 2006 09:12 Go to previous message
dev is currently offline  dev
Messages: 1693
Registered: September 2005
Location: Moscow
Senior Member

We have not seen any degradation here in real cases,
but probably you are right and pid hash can be allocated taking into account
physical memory as it is done for TCP/ip/other hashes?

But not sure, it is worth bothering right now... Maybe it worth first to make some
simple test, say:

1. run 50,000 tasks.
2. run some benchmark

and compare benchmark results with different hash sizes?
What do you think?

Kirill

> When looking at the linux pid hash table it is clearly tuned
> to a small number of processes < 4096. If there are more then
> that it starts to degrade with a worst case of 1024 entries
> on each hash chain when pid_max is pushed up to 4*1024*1024.
>
> If more process ids are common then it may be the case that
> we need to fix that data structure so it scales more gracefully.
>
> Eric
>
>
> ------------------------------------------------------------ -------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&b id=263057&dat=121642
> _______________________________________________
> Lxc-devel mailing list
> Lxc-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/lxc-devel
>
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: [Lxc-devel] New containers@lists.osdl.org mailing list
Next Topic: [PATCH 3/9] network namespaces: playing and debugging
Goto Forum:
  


Current Time: Sun Aug 03 16:09:05 GMT 2025

Total time taken to generate the page: 0.95222 seconds