[prev in list] [next in list] [prev in thread] [next in thread] 

List:       fuse-devel
Subject:    Re: [fuse-devel] fuse stat performance on multicore systems
From:       Han-Wen Nienhuys <hanwenn () gmail ! com>
Date:       2012-06-20 13:03:27
Message-ID: CAOw_e7a-8dx_p5BCVVZBbF0rGzixn+PskbM2WH-oC==4j8Q_5Q () mail ! gmail ! com
[Download RAW message or body]

On Wed, Jun 20, 2012 at 7:09 AM, Stefan Westerfeld <stefan@space.twc.de> wrote:

>> FUSE creates a thread per request, reusing threads if possible.
>> Another possible slowdown is that the FUSE threads run on different
>> CPUs and contend for shared locks to inspect the in-process inode
>> tree.
>
> I don't believe that can be the cause, because if there is a single userspace
> process that does stat/readdir calls, then only one FuSE thread should be
> active, as there is at most one open FuSE request to process at any given time.

if there is only one process active, there is no reason to move run
the FUSE process on a different CPU than the caller.

Anyway, measurements are king.

-- 
Han-Wen Nienhuys - hanwen@xs4all.nl - http://www.xs4all.nl/~hanwen

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
fuse-devel mailing list
fuse-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fuse-devel
[prev in list] [next in list] [prev in thread] [next in thread]