<p dir="ltr">6Gbps SAS. 12 in one array and 38 in another. It should saturate the bus.</p>
<div class="gmail_extra"><br><div class="gmail_quote">On Aug 20, 2016 8:41 PM, "DJ-Pfulio" <<a href="mailto:DJPfulio@jdpfu.com">DJPfulio@jdpfu.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Bus specifications have little to do with spinning disk performance, as<br>
you know.<br>
<br>
What is the max for each drive to read? 120-180Mbps? How are the<br>
arrays striped? 4-way, 8-way? That would mean at 4 x 120Mbps (480Mbps)<br>
is the highest possible throughput I'd expect. 960Mbps if 8-way stripe<br>
... there is overhead ... so 800Mbps seems reasonable to me.<br>
<br>
<a href="http://wintelguy.com/raidperf.pl" rel="noreferrer" target="_blank">http://wintelguy.com/raidperf.<wbr>pl</a> is more pessimistic on performance. I'm<br>
not convinced this numbers are MB/s. Think they are Mb/s.<br>
<br>
On 08/20/2016 06:49 PM, Ted W. wrote:<br>
> If you didn't already, try either iostat as `iostat -xm 1` or `sar -d`.<br>
> You see any unusual await or anything else that sticks out?<br>
><br>
> What FS? LVM?<br>
><br>
> A shot in the dark might lead me to check the FS alignment with fdisk but that<br>
> seems like an awfully large performance hit for something like that.<br>
><br>
> -Ted<br>
><br>
> On Fri, Aug 19, 2016 at 10:26:22AM -0400, Jim Kinney wrote:<br>
>> Getting DISMAL performance from new hardware.<br>
>><br>
>> Twin 6-core Xeon E5-2630L v.2 @2.40 GHz<br>
>> 128GB DDR4 RAM<br>
>> LSI megaraid 2108 in a PCIe v3 slot<br>
>> Onboard LSI 3108<br>
>> Drives are all 6Gbps SAS2 4TB<br>
>> 2 arrays, RAID 6, one is 40Tb, other 104TB<br>
>><br>
>> Everything hardware says I should expect a minimum through put of<br>
>> 12Gbps on either array.<br>
>><br>
>> The max I'm getting from iotop is 784 Mbps.<br>
>><br>
>> W. T. F!!!<br>
>><br>
>> According to top, the system is basically idle. Ditto from iostat and<br>
>> every other tool I check. I'm doing a 'cp -a' from/to same array.<br>
>><br>
>> Yes. New location is Luks encrypted. That process is totally asleep<br>
>> it's getting so little work. The rng is busy (dev/random is not running<br>
>> out (watch -n 1 cat /proc/sys/kernel/random/<wbr>entropy_avail was always<br>
>> above 2048)) but not overloaded or asleep.<br>
>><br>
>> Tested going unencrypted to unencrypted with similar performance.<br>
>><br>
>> Changed tuned-adm to latency-performance from balanced and the read<br>
>> portion of the cp went to 0 bps for 20 seconds until I switched it back<br>
>> to balanced mode. So that was a fail. It improved greatly using<br>
>> throughout-performance mode (bursts of 250MB/s) but was still showing<br>
>> long periods of 0 writes. Load went up from a paltry 8 to 22 so it's<br>
>> working more.<br>
>><br>
>> But top shows wa values across multiple cores hitting 100 so it's<br>
>> hitting a wall somewhere.<br>
>><br>
>> Any ideas of more places to look?<br>
><br>
>> ______________________________<wbr>_________________<br>
>> Ale mailing list<br>
>> <a href="mailto:Ale@ale.org">Ale@ale.org</a><br>
>> <a href="http://mail.ale.org/mailman/listinfo/ale" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo/ale</a><br>
>> See JOBS, ANNOUNCE and SCHOOLS lists at<br>
>> <a href="http://mail.ale.org/mailman/listinfo" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo</a><br>
><br>
> ______________________________<wbr>_________________<br>
> Ale mailing list<br>
> <a href="mailto:Ale@ale.org">Ale@ale.org</a><br>
> <a href="http://mail.ale.org/mailman/listinfo/ale" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo/ale</a><br>
> See JOBS, ANNOUNCE and SCHOOLS lists at<br>
> <a href="http://mail.ale.org/mailman/listinfo" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo</a><br>
><br>
<br>
<br>
--<br>
Got Linux? Used on smartphones, tablets, desktop computers, media<br>
centers, and servers by kids, Moms, Dads, grandparents and IT<br>
professionals.<br>
______________________________<wbr>_________________<br>
Ale mailing list<br>
<a href="mailto:Ale@ale.org">Ale@ale.org</a><br>
<a href="http://mail.ale.org/mailman/listinfo/ale" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo/ale</a><br>
See JOBS, ANNOUNCE and SCHOOLS lists at<br>
<a href="http://mail.ale.org/mailman/listinfo" rel="noreferrer" target="_blank">http://mail.ale.org/mailman/<wbr>listinfo</a><br>
</blockquote></div></div>