mfi driver performance too bad on LSI MegaRAID SAS 9260-8i

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

mfi driver performance too bad on LSI MegaRAID SAS 9260-8i

Jason Zhang
Hi,

I am working on storage service based on FreeBSD.  I look forward to a good result because many professional storage company use FreeBSD as its OS.  But I am disappointed with the Bad performance.  I tested the the performance of LSI MegaRAID 9260-8i and had the following bad result:

   1.  Test environment:
        (1) OS:   FreeBSD 10.0 release
        (2) Memory:  16G
        (3) RAID adapter:   LSI MegaRAID 9260-8i
        (4) Disks:  9 SAS hard drives (10000 rpm),  performance is expected for each hard drive
        (5) Test tools:   fio with  io-depth=1, thread num is 32 and block size is 64k or 1M
        (6)  RAID configuration:  RAID 5,   stripe size is 1M

  2.  Test result:
       (1)  write performance too bad:  20Mbytes/s throughput and 200 random write IOPS
       (2)  read performance is expected:  700Mbytes/s throughput and 1500 random read IOPS


I tested the same hardware configuration with CentOS linux and Linux's write performance is 5 times better than FreeBSD.      


Anyone encountered the same performance problem?  Does the mfi driver have performance issue or I should give up on FreeBSD?  





张京城   Jason

赛凡信息科技(厦门)有限公司              
Cyphy  Technology  (Xiamen)  Co.Ltd.
公司总部:厦门市软件园望海路55号A座901-904单元
研发总部:北京市东城区美术馆后街大取灯胡同2号
热线:4008798066
总机:0592-2936100
邮箱:[hidden email]
公司网址:Http://www.cyphytech.com


_______________________________________________
[hidden email] mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"
Reply | Threaded
Open this post in threaded view
|

Re: mfi driver performance too bad on LSI MegaRAID SAS 9260-8i

Julian Elischer-5
On 17/06/2016 3:16 PM, Jason Zhang wrote:

> Hi,
>
> I am working on storage service based on FreeBSD.  I look forward to a good result because many professional storage company use FreeBSD as its OS.  But I am disappointed with the Bad performance.  I tested the the performance of LSI MegaRAID 9260-8i and had the following bad result:
>
>     1.  Test environment:
>          (1) OS:   FreeBSD 10.0 release
>          (2) Memory:  16G
>          (3) RAID adapter:   LSI MegaRAID 9260-8i
>          (4) Disks:  9 SAS hard drives (10000 rpm),  performance is expected for each hard drive
>          (5) Test tools:   fio with  io-depth=1, thread num is 32 and block size is 64k or 1M
>          (6)  RAID configuration:  RAID 5,   stripe size is 1M
>
>    2.  Test result:
>         (1)  write performance too bad:  20Mbytes/s throughput and 200 random write IOPS
>         (2)  read performance is expected:  700Mbytes/s throughput and 1500 random read IOPS
>
>
> I tested the same hardware configuration with CentOS linux and Linux's write performance is 5 times better than FreeBSD.
>
>
> Anyone encountered the same performance problem?  Does the mfi driver have performance issue or I should give up on FreeBSD?
>
>
Unfortunatley issues related to performance can often be very specific.
We use the LSI cards with great success under FreeBSD 8 in our product
at work but it is impossible to say what is specifically wrong in your
setup.

Some years ago I did discover that fio needed to have completely
different arguments to get good performance under FreeBSD, so please
check that first.

What does performance look like with a single large write stream?

Also look at the handling of interrupts (systat -vmstat) to ensure
that interrupts are being handled correctly.
that can vary greatly from motherboard to motherboard  and bios to
bios. (even between revisions).
Sometimes Linux will cope differently  with these issues as they have
better support from the motherboard makers themselves.
(sometimes we cope better too).

One final thought.. make sure you have partitioned your drives and
filesyste,s so that all the block boundaries agree and line up.
At on place I worked we found we had accidentally  partitioned all our
drives starting 63 sectors into the drive.
That did NOT work well. :-) 8k raid stripe writes were always 2
writes  (and sometimes a read)

>
>
> 张京城   Jason
>
> 赛凡信息科技(厦门)有限公司
> Cyphy  Technology  (Xiamen)  Co.Ltd.
> 公司总部:厦门市软件园望海路55号A座901-904单元
> 研发总部:北京市东城区美术馆后街大取灯胡同2号
> 热线:4008798066
> 总机:0592-2936100
> 邮箱:[hidden email]
> 公司网址:Http://www.cyphytech.com
>
>
> _______________________________________________
> [hidden email] mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-performance
> To unsubscribe, send any mail to "[hidden email]"
>
>

_______________________________________________
[hidden email] mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"
Reply | Threaded
Open this post in threaded view
|

Re: mfi driver performance too bad on LSI MegaRAID SAS 9260-8i

Harry Schmalzbauer
In reply to this post by Jason Zhang
 Bezüglich Jason Zhang's Nachricht vom 17.06.2016 09:16 (localtime):
> Hi,
>
> I am working on storage service based on FreeBSD.  I look forward to a good result because many professional storage company use FreeBSD as its OS.  But I am disappointed with the Bad performance.  I tested the the performance of LSI MegaRAID 9260-8i and had the following bad result:
>
>    1.  Test environment:
>         (1) OS:   FreeBSD 10.0 release
>         (2) Memory:  16G
>         (3) RAID adapter:   LSI MegaRAID 9260-8i
>         (4) Disks:  9 SAS hard drives (10000 rpm),  performance is expected for each hard drive        

Were the drives completely initialized?
I remember that at least one vendor had implemented read-past-write for
every sector when written first.
It was with 15k 3.5" spindles and I'm really not sure which vendor it
was, so I won't name any. But "slow init" had solved a similir problem
for me back then...

-Harry

_______________________________________________
[hidden email] mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[hidden email]"