Messages in this thread | | | Subject | Re: SMP 2.1.131: SCSI performance extremely poor vs. IDE | From | Camm Maguire <> | Date | 14 Dec 1998 12:33:15 -0500 |
| |
Greetings! I'm running 2.1.130, SMP, 2 PII 350, 128 MB RAM, Adaptec 7890 U2 wide, 2 Seagate ST39173LW "Barracuda" drives. I patched the driver slightly so that the 80 MByte/s transfer speed is correctly negotiated. (I think you might have posted the patch?)
In any case, I'm not getting anything like these numbers. Basically, I don't see any noticeable improvement with RAID, except maybe for seeks. I haven't tried RAID-0 yet, just RAID-1, but I thought that could double the read performance. What does it mean to connect drives to different "channels" on the same controller? In any case, I thought the contoller had enough bandwidth to handle at least two drives full throttle. Any advice would be much appreciated.
Here are my bonnie results:
1 disk: -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 512 5905 95.1 11864 25.8 4372 15.0 6306 89.3 13744 18.5 117.9 2.0
RAID-1 partition: -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 512 5679 93.7 12624 33.6 4323 17.7 6272 90.2 14290 22.3 141.4 3.1
/proc/scsi/aic7xxx/0: Adaptec AIC7xxx driver version: 5.1.4/3.2.4 Compile Options: AIC7XXX_RESET_DELAY : 5 AIC7XXX_TAGGED_QUEUEING: Adapter Support Enabled Check below to see which devices use tagged queueing AIC7XXX_PAGE_ENABLE : Enabled (This is no longer an option) AIC7XXX_PROC_STATS : Disabled
Adapter Configuration: SCSI Adapter: Adaptec AIC-7890/1 Ultra2 SCSI host adapter Ultra2-LVD/SE Wide Controller PCI MMAPed I/O Base: 0xe0000000 Adapter SEEPROM Config: SEEPROM found and used. Adaptec SCSI BIOS: Enabled IRQ: 5 SCBs: Active 0, Max Active 2, Allocated 15, HW 32, Page 255 Interrupts: 518093 BIOS Control Word: 0x18a6 Adapter Control Word: 0x1c5e Extended Translation: Enabled Disconnect Enable Flags: 0xffff Ultra Enable Flags: 0x0000 Tag Queue Enable Flags: 0x0000 Ordered Queue Tag Flags: 0x0000 Default Tag Queue Depth: 8 Tagged Queue By Device array for aic7xxx host instance 0: {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} Actual queue depth per device for aic7xxx host instance 0: {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
Statistics: (scsi0:0:0:0) Device using Wide/Sync transfers at 80.0 MByte/sec, offset 15 Device Negotiation Settings Period Offset Bus Width User 010 127 1 Goal 010 127 1 Current 010 015 1 Total transfers 409840 (188177 read;221663 written) blks(512) rd=1275901; blks(512) wr=2719536
(scsi0:0:1:0) Device using Wide/Sync transfers at 80.0 MByte/sec, offset 15 Device Negotiation Settings Period Offset Bus Width User 010 127 1 Goal 010 127 1 Current 010 015 1 Total transfers 108173 (100970 read;7203 written) blks(512) rd=7221199; blks(512) wr=26402
/proc/scsi/scsi: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: SEAGATE Model: ST39173LW Rev: 5702 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 01 Lun: 00 Vendor: SEAGATE Model: ST39173LW Rev: 5702 Type: Direct-Access ANSI SCSI revision: 02
Thanks!
Doug Ledford <dledford@redhat.com> writes:
> Chris wrote: > > > > >>>>> "Lawrence" == Lawrence Walton <lawrence@otak-k.com> writes: > > > > Lawrence> I wonder about the bonnie benchmarking, but here is a > > Lawrence> SCSI benchmark with bonnie and a Symbios 875 and a 18 > > Lawrence> barracuda, P233 and 128 megs of ram. Linux version > > > > Lawrence> How does this stack up? is it a SMP only thing? > > > > Just as a side note, bonnie benchmarks of a size less than several > > times RAM, are just a test of memory bandwidth, since a large portion > > gets cached. Try comparing those numbers to a 800 - 1000 meg test... > > > > Heck, I've seen read rates of over 50meg/sec off of some small > > tests ;) If you want to see some benchmarks to compare against look > > at the bonnie results at > > > > http://www.tec.ualberta.ca/Benchmarks/benchmarks.html > > Those particular numbers all look to be from fairly dated hardware. > With somewhat modern hardware (5 Seagate Cheetah Ultra2 drives on two > Ultra2 channels on one 3950U2b Adaptec controller all in a RAID0 array > in a PII 266 box with 128MB RAM) I'm seeing numbers like these: > > 2.0.36UP: > -------Sequential Output-------- ---Sequential Input-- --Random-- > -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- > MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > 512 12020 98.2 65325 89.0 14788 50.0 14585 94.2 65283 86.7 255.1 5.5 > 512 12018 98.2 64312 87.8 14975 49.8 14707 95.3 65065 85.4 251.4 5.2 > 512 12041 98.2 66525 91.7 15027 50.2 14801 95.6 63949 86.2 258.8 5.6 > > The numbers I have for 2.1.131 so far are all SMP numbers and so can't > be compared against the 2.0.36 numbers. I plan on getting some UP > numbers for 2.1.131 as well before too long. > > -- > Doug Ledford <dledford@redhat.com> > Opinions expressed are my own, but > they should be everybody's. > > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.rutgers.edu > Please read the FAQ at http://www.tux.org/lkml/
-- Camm Maguire camm@enhanced.com ========================================================================== "The earth is but one country, and mankind its citizens." -- Baha'u'llah
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |