[Top] [All Lists]

Re: extremely slow file creation/deletion after xfs ran full

To: Stan Hoeppner <stan@xxxxxxxxxxxxxxxxx>
Subject: Re: extremely slow file creation/deletion after xfs ran full
From: Carsten Aulbert <Carsten.Aulbert@xxxxxxxxxx>
Date: Wed, 14 Jan 2015 07:07:00 +0100
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <54B58367.1070008@xxxxxxxxxxxxxxxxx>
Organization: Max Planck Institute for Gravitational Physics - Albert Einstein Institute (AEI)
References: <54B387A1.6000807@xxxxxxxxxx> <54B3CC6A.4080405@xxxxxxxxxx> <20150112155206.GD25944@xxxxxxxxxxxxxxx> <54B3F19D.6030307@xxxxxxxxxx> <20150112163749.GE25944@xxxxxxxxxxxxxxx> <54B40552.50106@xxxxxxxxxx> <54B57ACD.60600@xxxxxxxxxxxxxxxxx> <54B57C59.9070207@xxxxxxxxxx> <54B58367.1070008@xxxxxxxxxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.3.0
Hi Stan

On 01/13/2015 09:43 PM, Stan Hoeppner wrote:
> With your file sizes that seems a recipe for hotspots.  What do the
> controller metrics tell you about IOs per drive, bytes per drive?  Are
> they balanced?

They looked ok from 10000ft, but unfortunately, the controller is not
that talkative :(

>> After the arrival of "advanced format" HDD and SSDs we usually try to
>> align everything to full 1 MByte or larger, just to be sure any
>> combination of 512b, 4kb, ... will eventually align :)
> It's not that simple with striping.  Partitions need to start and end on
> stripe boundaries, not simply multiples of 4KB or 1MB as with single
> disks. If you use non power of 2 drive counts in a stripe, aligning at
> multiples of 1MB will screw ya, e.g. 12 drives * 64KB is a 768KB stripe.

At the moment there are 16 disks (or 8 "data" disks), so here, power of
2 still prevails. But still, very valid remark!



<Prev in Thread] Current Thread [Next in Thread>