Re: [patch v2 0/6] Add TRIM support for raid linear/0/1/10

[Roberto Spadim]

=/ well maybe with a sas disk it could be faster? maybe a pciexpress disk too?

Em 20 de março de 2012 23:08, Shaohua Li <shli@xxxxxxxxxx> escreveu:
> 2012/3/20 Shaohua Li <shli@xxxxxxxxxx>:
>> 2012/3/20 Holger Kiehl <Holger.Kiehl@xxxxxx>:
>>> Hello,
>>>
>>>
>>> On Tue, 20 Mar 2012, Shaohua Li wrote:
>>>
>>>> 2012/3/20 Holger Kiehl <Holger.Kiehl@xxxxxx>:
>>>>>
>>>>> Hello,
>>>>>
>>>>>
>>>>> On Fri, 16 Mar 2012, Shaohua Li wrote:
>>>>>
>>>>>> The patches add TRIM support for raid linear/0/1/10. I'll add TRIM
>>>>>> support
>>>>>> for
>>>>>> raid 4/5/6 later. The implementation is pretty straightforward and
>>>>>> self-explained.
>>>>>>
>>>>>> v1->v2:
>>>>>> 1. fixed a checking issue
>>>>>> 2. dropped discard request plug and replace it with no discard merege,
>>>>>> because
>>>>>> current SCSI layer can't handle discard request merge.
>>>>>>
>>>>> Have tested TRIM patches on three different systems with the following
>>>>> hardware/ setup:
>>>>>
>>>>>   1) root mounted on a raid1 over two SAS SSD's (200GB) and /home
>>>>> partition
>>>>>      on a raid0 over a fusionio ioDrive Duo. Is very new and seen very
>>>>>      little usage.
>>>>>
>>>>>   2) root and /home mounted on a raid0 over two Intel X25 Postville
>>>>>      (160GB) connected to a Intel P55 Express chipset. Has seen very
>>>>>      heavy usage for approx. 2 years.
>>>>>
>>>>>   3) root and /home mounted on a raid0 over three OCZ-VERTEX2 (120GB)
>>>>>      connected via ICH7 south bridge. Has seen mild usage for approx.
>>>>>      1.5 years.
>>>>>
>>>>> Made the following observations when running my own benchmark which
>>>>> copies around a lot of small files and deletes them. The benchmark on
>>>>> all systems was always run only on the /home partition ie. on a raid0.
>>>>>
>>>>> For system 1) there is hardly any measurable differnce whether discard
>>>>> is enabled or not (~29000 files per second).
>>>>>
>>>>> On system 2) the performance drops from 6500->3700 files per second,
>>>>> but under normal usage one does not notice any difference.
>>>>
>>>> do you have the blktrace data when the benchmark is running, especially
>>>> when doing file deletion. I'd like to check the latency of discard in this
>>>> case.
>>>>
>>> It is uploaded on ftp://ftp.dwd.de/pub/afd/test/trim
>> Thanks, I'll check it.
> Thanks for the testing. The trace data is very helpful. In the intel
> SSD, trace data
> shows a discard request uses about 1 ~ 3 ms. The filesystem suffers from
> fragmentation too, so lots of small discard requests. When ext4 starts doing
> discard, it usually uses more than 1 minutes. That's too bad.
> If just looking one disk's trace data, there are some extra latencies between
> two discard requests. The combined trace data of two disks show the latency
> comes from waiting for another disk, so nothing abnormal. I thought we could
> do an optimization for this case in the future.
> So in summary, discard from the SSDs is slow. When your filesystem is
> fragmented, the performance will be terrible.
>
> Thanks,
> Shaohua
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Roberto Spadim
Spadim Technology / SPAEmpresarial
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/