Re: HBA versus software initiator (open-iscsi)
On Fri, Jun 05, 2009 at 12:57:11AM +0200, benoit plessis wrote: Hi, Well the target are NetApp Filers, one FAS2020 and the other a FAS3020. The test system is directly attached to the filer, using either a bnx2 card or an intel The latency is calculated by bonnie++ 1.95 on a ext2 file system, the bandwith came from interface monitoring using cacti or bmon and also bonnie++ Iperf give 924Mbits/sec using iperf Using a 4k dd read on a nfs file mounted on the same filer i got 96Mo/s using an un-optimized path (no jumbo frame, going thru a few switchs and a router) Device blocksize is 4k, as is the FS blocksize OK. I don't really have personal experience with NetApp filers. Maybe they're good in NAS stuff, not in (iSCSI) SAN stuff? If you get 924 Mbit/sec with iperf using that bnx2 card, then the card and drivers should be fine. What distribution are you using on the initiator? What kernel? What open-iscsi version? Can you paste your iSCSI sessions parameters/settings from the active session? Did you try with 512 bytes device blocksize? -- Pasi 2009/6/4 Pasi Kärkkäinen pa...@iki.fi On Thu, Jun 04, 2009 at 04:03:28PM +0200, benoit plessis wrote: Hi, What do you think of (real) iSCSI HBA like Qlogic cards under linux ? We are using open-iscsi now with bnx2 cards and the perfs are far from decent (top 50/100Mbps), some comparative tests on an intel e1000 card show betters results (500/600Mbps) but far from gigabit saturation, and still high latency. What target are you using? nullio mode or memdisk is good for benchmarking, if either of those are possible on your target. How did you measure the latency? Are you using direct crossover cables, or switches? How does your NICs perform with for example FTP? How about iperf? How did you get those 50/100 and 500/600 Mbit numbers? What benchmark did you use? What kind of blocksize? -- Pasi --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: HBA versus software initiator (open-iscsi)
2009/6/5 Pasi Kärkkäinen pa...@iki.fi On Fri, Jun 05, 2009 at 12:57:11AM +0200, benoit plessis wrote: Hi, Well the target are NetApp Filers, one FAS2020 and the other a FAS3020. The test system is directly attached to the filer, using either a bnx2 card or an intel The latency is calculated by bonnie++ 1.95 on a ext2 file system, the bandwith came from interface monitoring using cacti or bmon and also bonnie++ Iperf give 924Mbits/sec using iperf Using a 4k dd read on a nfs file mounted on the same filer i got 96Mo/s using an un-optimized path (no jumbo frame, going thru a few switchs and a router) Device blocksize is 4k, as is the FS blocksize OK. I don't really have personal experience with NetApp filers. Maybe they're good in NAS stuff, not in (iSCSI) SAN stuff? When i told them my number they seem very surprised and said they do have better results ... If you get 924 Mbit/sec with iperf using that bnx2 card, then the card and drivers should be fine. As of the card, while i can have very good result of this kind, i know by experience that in some workload conditions bnx2 cards simply fail. I have similar system that can't get past 60Mbps of http traffic using haproxy. What distribution are you using on the initiator? What kernel? What open-iscsi version? Distribution is debian/lenny, kernel 2.6.26-2-amd64 and packaged open-iscsi 2.0.870~rc3-0.4 Can you paste your iSCSI sessions parameters/settings from the active session? Well yes, if you told me how or where to look :) Did you try with 512 bytes device blocksize I just can't: r...@zeus:~# blockdev --setbsz 512 /dev/sdb r...@zeus:~# blockdev --getbsz /dev/sdb 4096 r...@zeus:~# blockdev --setbsz 8192 /dev/sdb BLKBSZSET: Argument invalide --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: HBA versus software initiator (open-iscsi)
Looks like you are right, some quick tests made on a iscsitarget linux server using nullio and fileio on a sata drive give much better results than the netapp file server with it's 12 disks (and while the netapp is directly connected to the client and on a jumbo frame network, the linux initiator is contacted thru 2 vlans and one router in standard frames ...) netapp: r...@zeus:~# dd if=/dev/zero of=/dev/sdb bs=4096 count=1251481 1251481+0 enregistrements lus 1251481+0 enregistrements écrits 5126066176 bytes (5,1 GB) copied, 192,076 s, 26,7 MB/s linux nullio: r...@zeus:~# dd if=/dev/zero of=/dev/sdc bs=4096 count=1251481 1251481+0 enregistrements lus 1251481+0 enregistrements écrits 5126066176 bytes (5,1 GB) copied, 45,5091 s, 113 MB/s linux fileio: r...@zeus:~# dd if=/dev/zero of=/dev/sdc bs=4096 count=1251481 1251481+0 enregistrements lus 1251481+0 enregistrements écrits 5126066176 bytes (5,1 GB) copied, 114,309 s, 44,8 MB/s --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---