|To:||Dave Chinner <david@xxxxxxxxxxxxx>|
|Subject:||Re: concurrent direct IO write in xfs|
|From:||Zheng Da <zhengda1936@xxxxxxxxx>|
|Date:||Mon, 23 Jan 2012 14:34:34 -0500|
|Dkim-signature:||v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=cmknQ0G8nYG21L1NePPEVi20ifWhlcGDB0QT79V3UhQ=; b=uHz8iN0c7D+i6iY6uXvkEO0Q0M0fLhx6dVw5I1MT2IBUsgU0gSqpqKAkJvAnsoAlNa PcXvukhDdFZdIY31X0Ngz5eChKVjUT1FEiI8jqZzz8wBYMFij+c/oQWPGoZRvMqgOeEI vbofaqkHEckpu3+eSMqQkGx7VraZEF7ud72gU=|
|References:||<CAFLer83FBZG9ZCrT2jUZBcTC2a2tx_CDmykyPF4cTP0dbHGw7Q@xxxxxxxxxxxxxx> <20120116232549.GC6922@dastard> <CAFLer81XkMTh_gxd95pzxCEs1yGRsTrZijX3c7ewgRzeA7DCSQ@xxxxxxxxxxxxxx> <20120123051155.GI15102@dastard>|
On Mon, Jan 23, 2012 at 12:11 AM, Dave Chinner <david@xxxxxxxxxxxxx> wrote:
Here I was just wondering if i_new_size is always 0 if there are only overwrites. I think it has nothing to do with the pattern of my workloads or the device I used for the test.
It's difficult to test the scalability problem in the traditional disks. They provide very low IOPS (IO per second). Even two SSDs can't provide enough IOPS.
I don't think all direct IO will serialized on the extent tree lookups. Direct IO reads can parallelized pretty well and they also need extent tree lookups.
OK, I can try that.
I can do that. My test code gets very complicated now. I need to simplify it.
|<Prev in Thread]||Current Thread||[Next in Thread>|
|Previous by Date:||Re: [patch 04/12] xfs: cleanup xfs_log_space_wake, Mark Tinguely|
|Next by Date:||Re: [PATCH] xfsprogs: check for size parsing errors in xfs_quota, Christoph Hellwig|
|Previous by Thread:||Re: concurrent direct IO write in xfs, Dave Chinner|
|Next by Thread:||Re: concurrent direct IO write in xfs, Zheng Da|
|Indexes:||[Date] [Thread] [Top] [All Lists]|