[OM Cooker] Problem: SSD drastic drop in hdparm/disk reads.

Ben Bullard benbullard79 at cox.net
Mon Dec 26 20:18:56 EST 2016


Thanks for your help. I will read the manual on tuned.


Ben Bullard
ben79
--------------------
OpenMandriva-QA Team

On 12/26/2016 07:00 PM, Tomasz Gajc wrote:
> Great then. Please be aware that tuned is doing a lot optimizations. 
> In my case it was setting APM to low on sata links which reduced 
> performance. Anyways tuned can be adjusted with proper settings. Read 
> manual.
>
> 27.12.2016 12:28 AM "Ben Bullard" <benbullard79 at cox.net 
> <mailto:benbullard79 at cox.net>> napisał(a):
>
>     I tried throughput-performance, desktop, and realtime and had same
>     problem with all. Then I tried disabling tuned.service and that
>     worked in that there are no longer drops or fluctuations in read
>     speed under hdparm over 10 hour time period.
>
>     Ben Bullard
>     ben79
>     --------------------
>     OpenMandriva-QA Team
>
>     On 12/25/2016 05:19 PM, Tomasz Paweł Gajc wrote:
>>     Dnia niedziela, 25 grudnia 2016 16:58:19 CET Ben Bullard pisze:
>>>     Drive in question: Model Family:     Crucial/Micron
>>>     MX100/MX200/M5x0/M600 Client SSDs
>>>
>>>     Device Model:     Crucial_CT500MX200SSD4
>>>
>>>     Firmware Version: MU04
>>>
>>>     Ok, this problem isn't going away. TPG, Colin, Bero, or anyone what logs
>>>     or test/commands in addition to omv-bug-report.log do I need to include
>>>     for a bug report on this.
>>     Hard to say what may be the problem.
>>     Maybe it's related to aggresive APM on your SATA ports.
>>
>>     tuned daemon may interfere here.
>>
>>     take a look on:
>>     tuned-adm active
>>
>>     then change to diffrent profile:
>>
>>     tuned-adm profile throughput-performance
>>
>>     and then run hdparm tests
>>
>>>     There was a kernel-firmware (20161225) update
>>>     today I hoped would address this but it didn't.
>>     kernel-firmware has nothing to do with firmware with your disk.
>>
>>
>>     On my SSD Samsung EVO 840 on Lx 3.0 i got stable results:
>>
>>     [root at lazur tpg]# hdparm -t /dev/sda
>>
>>     /dev/sda:
>>       Timing buffered disk reads: 762 MB in  3.01 seconds = 253.56 MB/sec
>>     [root at lazur tpg]# hdparm -t /dev/sda
>>
>>     /dev/sda:
>>       Timing buffered disk reads: 778 MB in  3.00 seconds = 259.29 MB/sec
>>     [root at lazur tpg]# hdparm -t /dev/sda
>>
>>     /dev/sda:
>>       Timing buffered disk reads: 778 MB in  3.01 seconds = 258.63 MB/sec
>>
>>
>>
>>
>>     _______________________________________________
>>     OM-Cooker mailing list
>>     OM-Cooker at ml.openmandriva.org <mailto:OM-Cooker at ml.openmandriva.org>
>>     http://ml.openmandriva.org/mailman/listinfo/om-cooker_ml.openmandriva.org
>>     <http://ml.openmandriva.org/mailman/listinfo/om-cooker_ml.openmandriva.org>
>     _______________________________________________ OM-Cooker mailing
>     list OM-Cooker at ml.openmandriva.org
>     <mailto:OM-Cooker at ml.openmandriva.org>
>     http://ml.openmandriva.org/mailman/listinfo/om-cooker_ml.openmandriva.org
>     <http://ml.openmandriva.org/mailman/listinfo/om-cooker_ml.openmandriva.org>
>
>
> _______________________________________________
> OM-Cooker mailing list
> OM-Cooker at ml.openmandriva.org
> http://ml.openmandriva.org/mailman/listinfo/om-cooker_ml.openmandriva.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ml.openmandriva.org/pipermail/om-cooker_ml.openmandriva.org/attachments/20161226/84121f66/attachment-0001.html>


More information about the OM-Cooker mailing list