How to speed optimize SATA RAID5 implemented with Lycom/JMicron card
Linux - HardwareThis forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
How to speed optimize SATA RAID5 implemented with Lycom/JMicron card
I'm using a Lycom ST-158 card ( JMicron chip, here http://www.lycom.com.tw/ST158.htm ) to implement a RAID5 on an embedded system. The card is attached directly to SATA controller on SoC. Libata in kernel 2.6.
What can I do to optimize speed - especially read?
1) It appears to properly be doing SATA II ( 3 Gb )
2) I made ext3 fs with 8-sector alignment
3) read-lookahead didnt seem to help much
4) hdparm reports logical & phys sector size of 512 though the WD disks are 4096. I guess the card isnt transparent for 4096?
Yes - on 1GB files R/W from RAID( read from /dev/zero or written to /dev/null ) - I'm getting usually about 107MB/s read, 72 write. Its maybe 10% higher going straight to one of those disks without the RAID card ( which is in RAID5 mode using 3 disks ). But it does seem to vary across runs quite a bit.
Bah - if you get 250MB/s with an SSD and ~120MB/s when using 1 HDD, I would expect you to get ~200 when using the raid5
When you read from the raid5 and have a look with "htop" and/or "nmon" at the CPU usage, does it go up a lot?
Btw., you're using the Linux software-RAID, right?
Sorry - the system without the lycom card , just SSD, did ~250;
the system with 'normal' drive ~120
Whenever I did raid I used the CARD to to rAID , not linux SW raid. The system is ignorant of any RAID capability - it just sees the 3 drives as one, with size = 2x of one drive.
System isnt up right now but I'll check CPU etc, but I doubt anything unusual since there's no system knowledge of RAID.
Do you want to try perhaps to benchmark the SSD when connected to the card (in a non-raid configuration of course) to exclude that the card itself slows down the transfer?
Still have a look at the CPU, you never know - perhaps your HW-card implements RAID5 in software in the drives of the card itself => if the driver is not as good as the usual/normal SW-raid of the Linux kernel that might have an impact... (lazy - I did not check your card in the Internet).
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.