[Mono-list] Same code, different performance?

Rodrigo Kumpera kumpera at gmail.com
Fri Feb 27 16:23:34 EST 2009


Please fill a bug report with a small test case that shows the difference.
You can use the profiler we ship with mono to find the hotspots and point
them
on the bugreport.


On Thu, Feb 12, 2009 at 7:34 PM, reubensm <reubens73 at gmail.com> wrote:

>
> Hello Mono community :)
>
> At work we have a piece of C# code written and compiled under Windows that
> we'd like to port to Linux using Mono.
>
> The code has two parts (client and server) that perform many tasks, one of
> which is data transfer from the cilent to the server.
>
> In our tests we see that when running on Windows, the same code achieves
> performance at least 3-4 times better (faster) than when running under
> Mono/Linux.
>
> We ran a profiler, but the results didn't give any obvious culprit, except
> the possibility that it might be related to synchronization primitives
> implementation of Mono.
>
> Can anyone think of a reason for this slowness? like said, it is the same
> code (and binaries, compiled under Windows) used in both Windows and Linux.
> Both environments are physical (no virtualization, VMs).
>
> Could this be unrelated to Mono, and a result of some OS settings? (Fedora
> 10) if so, what should we check?
>
> Thanks
>
> --
> View this message in context:
> http://www.nabble.com/Same-code%2C-different-performance--tp21986473p21986473.html
> Sent from the Mono - General mailing list archive at Nabble.com.
>
> _______________________________________________
> Mono-list maillist  -  Mono-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-list
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ximian.com/pipermail/mono-list/attachments/20090227/68e44bb9/attachment.html 


More information about the Mono-list mailing list