Re: MegaRAID SCSI 320-2 bad write performance
For comparison with the SATA 150-6: [EMAIL PROTECTED]:/home/ethant# bioctl -i ami0 Volume Status Size Device ami0 0 Online 107374182400 sd0 RAID1 0 Online 160036814848 0:0.0 noencl ST3160811AS 3.AA 1 Online 160036814848 0:1.0 noencl ST3160811AS 3.AA ami0 1 Online42949672960 sd1 RAID1 0 Online 160036814848 0:0.0 noencl ST3160811AS 3.AA 1 Online 160036814848 0:1.0 noencl ST3160811AS 3.AA ami0 2 Online 9712959488 sd2 RAID1 0 Online 160036814848 0:0.0 noencl ST3160811AS 3.AA 1 Online 160036814848 0:1.0 noencl ST3160811AS 3.AA ami0 3 Hot spare160036814848 0:2.0 noencl ST3160811AS 3.AA [EMAIL PROTECTED]:/home/ethant# dd if=/dev/zero of=/dev/rsd1c bs=64k count=4096 4096+0 records in 4096+0 records out 268435456 bytes transferred in 6.441 secs (41674946 bytes/sec) OpenBSD 4.0-beta (GENERIC) #1083: Mon Aug 21 21:24:02 MDT 2006 [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC cpu0: AMD Athlon(tm) XP 1800+ (AuthenticAMD 686-class, 256KB L2 cache) 1.54 GHz cpu0:FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA, CMOV,PAT,PSE36,MMX,FXSR,SSE real mem = 267939840 (261660K) avail mem = 236683264 (231136K) On 8/22/06, Marco Peereboom [EMAIL PROTECTED] wrote: Due to the battery missing every IO the host sends has to complete before the next one goes down. So the sequence of events is: 1. Send host io through driver 2. Firmware accepts it 3. Firmware creates 1 or more IOs and shoots those off to the disk 4. Firmware waits until IOs complete 5. Firmware raises interrupt to inform host that IO completed 6. ami(4) driver now completes the IO on the host 7. goto 1 Also if you want to test read/write performance you have to use the raw device. A dd test should use for example /dev/rsd0c instead of a file. Why don't you have a battery for that thing? On a separate note; i do believe that LSI does have some firmware that'll allow to enable write back cache without a battery. I am not 100% sure about though. On Wed, Aug 23, 2006 at 04:01:10AM +0200, Robert Urban wrote: below are the results of my tests with the LSI MegaRAID SCSI 320-2 controller on 3.9-release and 4.0-beta (snapshot pulled on Aug 22, 2006). I tested writing to both a RAID-5 and a RAID-0 logical drive. The RAID-0 drive consists of a single drive, what I'd call a JBOD. While the test program was running, I had iostat running, and I noted the results next to each test. The iostat values moved around a lot, so I took a figure close to the peak value. Naturally all the async tests were influenced by the buffer cache to a certain extent. I'm a little confused as to why writing sequential blocks with O_SYNC should be so slow...
Re: MegaRAID SCSI 320-2 bad write performance
I set up a couple of partitions to perform raw I/O to, one on the RAID-5 drive and one on the RAID-0 drive: dd to raw device: RAID-5: 6.4MB/s RAID-0: 9.1MB/s There is *no* difference between 3.9 and 4.0-beta. For kicks, I also tested reading from a raw device: dd reading from a raw device: RAID-5: 38.3MB/s RAID-0: 81.4MB/s For what it's worth, the numbers for writing and reading are identical under linux too. Summary: hw-raid controllers are slow for writing, at least without a cache battery. last question: what kind of performance can I expect if I get the battery? Rob Urban
Re: MegaRAID SCSI 320-2 bad write performance
On 8/23/06, Robert Urban [EMAIL PROTECTED] wrote: I set up a couple of partitions to perform raw I/O to, one on the RAID-5 drive and one on the RAID-0 drive: dd to raw device: RAID-5: 6.4MB/s RAID-0: 9.1MB/s Hmmm, if I'm doing my calculations right I'm getting almost 40MB/s for RAID-1 which should be slower than RAID-0, shouldn't it? There is *no* difference between 3.9 and 4.0-beta. For kicks, I also tested reading from a raw device: dd reading from a raw device: RAID-5: 38.3MB/s RAID-0: 81.4MB/s For what it's worth, the numbers for writing and reading are identical under linux too. Summary: hw-raid controllers are slow for writing, at least without a cache battery. last question: what kind of performance can I expect if I get the battery? When I get back from my vacation to China I think I'll get the battery for my 150-6. Greg
Re: MegaRAID SCSI 320-2 bad write performance
Greg wrote: Hmmm, if I'm doing my calculations right I'm getting almost 40MB/s for RAID-1 which should be slower than RAID-0, shouldn't it? I'm not sure what you mean by my calculations. In your earlier mail you wrote: [EMAIL PROTECTED]:/home/ethant# dd if=/dev/zero of=/dev/rsd1c bs=64k count=4096 4096+0 records in 4096+0 records out 268435456 bytes transferred in 6.441 secs (41674946 bytes/sec) dd is kind enough to tell you the b/w :) I guess you mean this. RAID-1, mirroring, is *not necessarily* slower than RAID-0, striping. Somewhere, something has to issue two write operations, one to each mirror, and then wait for both to finish before announcing that the I/O is finished. This may take a bit more time compared to a single disk, but it shouldn't be much. If you have multiple disks in a RAID-0 array, that should be much faster, of course. Rob Urban
Re: MegaRAID SCSI 320-2 bad write performance
On Aug 22, 2006, at 2:17 AM, Robert Urban wrote: Hi Folks, using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive consisting of a RAID-5 set of 4 72GB HP Ultra320 disks and to a RAID-0 drive consisting of a single 300GB HP Ultra320 disk. The controller has 128MB of cache, but I do not have a battery, so cache write policy is write-through. The kernel is the bsd, and not bsd.mp, but it makes no difference. I've tried both. Had the same kind of low performance on a Dell server with an entry level RAID adapter. I took the risk to force cache write policy to write-back, even without a battery. Performance is as it should be for a RAID-0, I'm just sweating a bit more... Next time, will add a few bucks and buy a real controller. Regards.
Re: MegaRAID SCSI 320-2 bad write performance
below are the results of my tests with the LSI MegaRAID SCSI 320-2 controller on 3.9-release and 4.0-beta (snapshot pulled on Aug 22, 2006). I tested writing to both a RAID-5 and a RAID-0 logical drive. The RAID-0 drive consists of a single drive, what I'd call a JBOD. While the test program was running, I had iostat running, and I noted the results next to each test. The iostat values moved around a lot, so I took a figure close to the peak value. Naturally all the async tests were influenced by the buffer cache to a certain extent. I'm a little confused as to why writing sequential blocks with O_SYNC should be so slow... Rob Urban #== # 3.9 #== RAID-0 drive: write-test: -- Sync -- 1 proc: 1.7MB/s iostat: 1.8MB/s 2 procs: 1MB/s and 1MB/s iostat: 2MB/s Async -- 1 proc: 8.7MB/s iostat: 5MB/s 2 procs: 11MB/s and 5.6MB/s iostat: 4.8MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 28.26 seconds, bw: 9MB/s iostat reported: 4.6MB/s RAID-5 drive: write-test: -- Sync -- 1 proc: 1.1MB/s iostat: 1.2MB/s 2 procs: 0.7MB/s and 0.7MB/s iostat: 1.6MB/s Async -- 1 proc: 10MB/s iostat: 3.2MB/s 2 procs: 5.2MB/s and 5.2MB/s iostat: 5MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 39.1 seconds, bw: 6.5MB/s iostat reported: 3.3MB/s #== # 4.0-beta #== RAID-0 drive: write-test: -- Sync -- 1 proc: 1.5MB/s iostat: 3MB/s 2 procs: 0.76MB/s and 0.76MB/s iostat: 3.4MB/s Async -- 1 proc: 15MB/s iostat: 10MB/s 2 procs: 8.2MB/s and 5.8MB/s iostat: 10MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 21.8 seconds, bw: 11.7MB/s iostat reported: 10MB/s RAID-5 drive: write-test: -- Sync -- 1 proc: 1.02MB/s iostat: 2.5MB/s 2 procs: 0.62MB/s and 0.62MB/s iostat: 2.9MB/s Async -- 1 proc: 9MB/s iostat: 6.3MB/s 2 procs: 4.1MB/s and 3.8MB/s iostat: 6.5MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 35 seconds, bw: 7.3MB/s iostat reported: 7MB/s
Re: MegaRAID SCSI 320-2 bad write performance
On 8/21/06, Robert Urban [EMAIL PROTECTED] wrote: using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive . . . All tests performed with O_SYNC, to avoid bufcache interaction. The performance, at least to my perhaps naive eyes, seems abysmal. I'm getting 1.2MB/sec on the RAID-5 logical drive. Have you tried your test on any other controller? Do you have bonnie++ results for this controller? I have a pair of Dell 2850 servers with PERC 4e/Di, if these are similar enough I can run tests (RAID0 and RAID1). Kevin
Re: MegaRAID SCSI 320-2 bad write performance
Hi Kevin, Kevin wrote: On 8/21/06, Robert Urban [EMAIL PROTECTED] wrote: using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive . . . All tests performed with O_SYNC, to avoid bufcache interaction. The performance, at least to my perhaps naive eyes, seems abysmal. I'm getting 1.2MB/sec on the RAID-5 logical drive. Have you tried your test on any other controller? no. I could run them on the onboard Smart Array 5i controller. I will try to get some results for this controller, maybe tomorrow. Do you have bonnie++ results for this controller? No, but I've just copied the sources :) Will post. In order to get a baseline, I could plug in an AHA-3960D (dual channel U160) and connect a disk to it, just to see what the disk can do with no raid controller in between... I have a pair of Dell 2850 servers with PERC 4e/Di, if these are similar enough I can run tests (RAID0 and RAID1). with cache battery back up? Rob Urban
Re: MegaRAID SCSI 320-2 bad write performance
Due to the battery missing every IO the host sends has to complete before the next one goes down. So the sequence of events is: 1. Send host io through driver 2. Firmware accepts it 3. Firmware creates 1 or more IOs and shoots those off to the disk 4. Firmware waits until IOs complete 5. Firmware raises interrupt to inform host that IO completed 6. ami(4) driver now completes the IO on the host 7. goto 1 Also if you want to test read/write performance you have to use the raw device. A dd test should use for example /dev/rsd0c instead of a file. Why don't you have a battery for that thing? On a separate note; i do believe that LSI does have some firmware that'll allow to enable write back cache without a battery. I am not 100% sure about though. On Wed, Aug 23, 2006 at 04:01:10AM +0200, Robert Urban wrote: below are the results of my tests with the LSI MegaRAID SCSI 320-2 controller on 3.9-release and 4.0-beta (snapshot pulled on Aug 22, 2006). I tested writing to both a RAID-5 and a RAID-0 logical drive. The RAID-0 drive consists of a single drive, what I'd call a JBOD. While the test program was running, I had iostat running, and I noted the results next to each test. The iostat values moved around a lot, so I took a figure close to the peak value. Naturally all the async tests were influenced by the buffer cache to a certain extent. I'm a little confused as to why writing sequential blocks with O_SYNC should be so slow... Rob Urban #== # 3.9 #== RAID-0 drive: write-test: -- Sync -- 1 proc: 1.7MB/s iostat: 1.8MB/s 2 procs: 1MB/s and 1MB/s iostat: 2MB/s Async -- 1 proc: 8.7MB/s iostat: 5MB/s 2 procs: 11MB/s and 5.6MB/s iostat: 4.8MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 28.26 seconds, bw: 9MB/s iostat reported: 4.6MB/s RAID-5 drive: write-test: -- Sync -- 1 proc: 1.1MB/s iostat: 1.2MB/s 2 procs: 0.7MB/s and 0.7MB/s iostat: 1.6MB/s Async -- 1 proc: 10MB/s iostat: 3.2MB/s 2 procs: 5.2MB/s and 5.2MB/s iostat: 5MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 39.1 seconds, bw: 6.5MB/s iostat reported: 3.3MB/s #== # 4.0-beta #== RAID-0 drive: write-test: -- Sync -- 1 proc: 1.5MB/s iostat: 3MB/s 2 procs: 0.76MB/s and 0.76MB/s iostat: 3.4MB/s Async -- 1 proc: 15MB/s iostat: 10MB/s 2 procs: 8.2MB/s and 5.8MB/s iostat: 10MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 21.8 seconds, bw: 11.7MB/s iostat reported: 10MB/s RAID-5 drive: write-test: -- Sync -- 1 proc: 1.02MB/s iostat: 2.5MB/s 2 procs: 0.62MB/s and 0.62MB/s iostat: 2.9MB/s Async -- 1 proc: 9MB/s iostat: 6.3MB/s 2 procs: 4.1MB/s and 3.8MB/s iostat: 6.5MB/s dd if=/dev/zero of=testfile bs=64k count=4096 -- took 35 seconds, bw: 7.3MB/s iostat reported: 7MB/s
MegaRAID SCSI 320-2 bad write performance
Hi Folks, using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive consisting of a RAID-5 set of 4 72GB HP Ultra320 disks and to a RAID-0 drive consisting of a single 300GB HP Ultra320 disk. The controller has 128MB of cache, but I do not have a battery, so cache write policy is write-through. The kernel is the bsd, and not bsd.mp, but it makes no difference. I've tried both. All tests performed with O_SYNC, to avoid bufcache interaction. The performance, at least to my perhaps naive eyes, seems abysmal. I'm getting 1.2MB/sec on the RAID-5 logical drive. It doesn't matter what blocksize I choose, 32k, 64k, 128k, 256k. The chunksize on the logical drive is the default, 64k. On the single HP 300GB Ultra320 RAID-0 drive write performance is also bad: 1.7MB/sec. The systems is (in case you hadn't already guessed) a Compaq ProLiant DL380-G2 with 1GB main memory and dual 1266MHz CPUs. I will attach dmesg, my test program, and the output of bioctl -iv ami0 below. I can't imagine the MegaRAID controller is this slow. Any suggestions? Rob Urban dmesg: --- OpenBSD 3.9 (GENERIC) #617: Thu Mar 2 02:26:48 MST 2006 [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC cpu0: Intel(R) Pentium(R) III CPU family 1266MHz (GenuineIntel 686-class) 1.27 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,M MX,FXSR,SSE real mem = 1073307648 (1048152K) avail mem = 972660736 (949864K) using 4278 buffers containing 53768192 bytes (52508K) of memory mainbus0 (root) bios0 at mainbus0: AT/286+(00) BIOS, date 12/31/99, BIOS32 rev. 0 @ 0xf pcibios0 at bios0: rev 2.1 @ 0xf/0x2000 pcibios0: PCI BIOS has 9 Interrupt Routing table entries pcibios0: PCI Interrupt Router at 000:15:0 (ServerWorks OSB4 rev 0x00) pcibios0: PCI bus #0 is the last bus bios0: ROM list: 0xc/0x8000 0xc8000/0x2600 0xee000/0x2000! cpu0 at mainbus0 pci0 at mainbus0 bus 0: configuration mode 1 (no bios) pchb0 at pci0 dev 0 function 0 ServerWorks CNB20HE Host rev 0x23 pci1 at pchb0 bus 1 ppb0 at pci1 dev 3 function 0 Intel i960 RP PCI-PCI rev 0x05 pci2 at ppb0 bus 2 vga1 at pci2 dev 0 function 0 ATI Mach64 GV rev 0x7a wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation) wsdisplay0: screen 1-5 added (80x25, vt100 emulation) Intel 80960RP ATU rev 0x05 at pci1 dev 3 function 1 not configured pchb1 at pci0 dev 0 function 1 ServerWorks CNB20HE Host rev 0x01 pchb2 at pci0 dev 0 function 2 ServerWorks CNB20HE Host rev 0x01 pchb3 at pci0 dev 0 function 3 ServerWorks CNB20HE Host rev 0x01 pci3 at pchb3 bus 7 ami0 at pci3 dev 4 function 0 Symbios Logic MegaRAID rev 0x01: irq 3 Dell 518 64b/lhc ami0: FW 351X, BIOS v1.10, 128MB RAM ami0: 2 channels, 0 FC loops, 2 logical drives scsibus0 at ami0: 40 targets sd0 at scsibus0 targ 0 lun 0: AMI, Host drive #00, SCSI2 0/direct fixed sd0: 208110MB, 208110 cyl, 64 head, 32 sec, 512 bytes/sec, 426209280 sec total sd1 at scsibus0 targ 1 lun 0: AMI, Host drive #01, SCSI2 0/direct fixed sd1: 286080MB, 286080 cyl, 64 head, 32 sec, 512 bytes/sec, 585891840 sec total scsibus1 at ami0: 16 targets uk0 at scsibus1 targ 15 lun 0: COMPAQ, PROLIANT 4L6I, 1.84 SCSI2 3/processor fixed uk0: unknown device scsibus2 at ami0: 16 targets Compaq PCI Hotplug rev 0x12 at pci3 dev 7 function 0 not configured fxp0 at pci0 dev 2 function 0 Intel 8255x rev 0x08, i82559: irq 5, address 00:08:02:8a:4b:fc inphy0 at fxp0 phy 1: i82555 10/100 PHY, rev. 4 fxp1 at pci0 dev 4 function 0 Intel 8255x rev 0x08, i82559: irq 7, address 00:08:02:8a:4b:fb inphy1 at fxp1 phy 1: i82555 10/100 PHY, rev. 4 Compaq Netelligent ASMC rev 0x00 at pci0 dev 6 function 0 not configured piixpm0 at pci0 dev 15 function 0 ServerWorks OSB4 rev 0x51: SMBus disabled pciide0 at pci0 dev 15 function 1 ServerWorks OSB4 IDE rev 0x00: DMA atapiscsi0 at pciide0 channel 0 drive 0 scsibus3 at atapiscsi0: 2 targets cd0 at scsibus3 targ 0 lun 0: COMPAQ, CD-ROM SN-124, N102 SCSI0 5/cdrom removable cd0(pciide0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 1 ohci0 at pci0 dev 15 function 2 ServerWorks OSB4/CSB5 USB rev 0x04: irq 11, version 1.0, legacy support usb0 at ohci0: USB revision 1.0 uhub0 at usb0 uhub0: ServerWorks OHCI root hub, rev 1.00/1.00, addr 1 uhub0: 4 ports with 4 removable, self powered isa0 at mainbus0 isadma0 at isa0 pckbc0 at isa0 port 0x60/5 pckbd0 at pckbc0 (kbd slot) pckbc0: using irq 1 for kbd slot wskbd0 at pckbd0: console keyboard, using wsdisplay0 pms0 at pckbc0 (aux slot) pckbc0: using irq 12 for aux slot wsmouse0 at pms0 mux 0 pcppi0 at isa0 port 0x61 midi0 at pcppi0: PC speaker spkr0 at pcppi0 npx0 at isa0 port 0xf0/16: using exception 16 pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo fdc0 at isa0 port 0x3f0/6 irq 6 drq 2 fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec biomask ef4d netmask efed ttymask ffef pctr: 686-class
Re: MegaRAID SCSI 320-2 bad write performance
On 22/08/2006, at 10:17 AM, Robert Urban wrote: Hi Folks, using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive consisting of a RAID-5 set of 4 72GB HP Ultra320 disks and to a RAID-0 drive consisting of a single 300GB HP Ultra320 disk. The controller has 128MB of cache, but I do not have a battery, so cache write policy is write-through. The kernel is the bsd, and not bsd.mp, but it makes no difference. I've tried both. bsd should be slightly faster than bsd.mp, but not enough for you to notice. Can you bring the box up to a snapshot? All tests performed with O_SYNC, to avoid bufcache interaction. if you really want to avoid that then use /dev/rsdXX. The performance, at least to my perhaps naive eyes, seems abysmal. I'm getting 1.2MB/sec on the RAID-5 logical drive. It doesn't matter what blocksize I choose, 32k, 64k, 128k, 256k. The chunksize on the logical drive is the default, 64k. On the single HP 300GB Ultra320 RAID-0 drive write performance is also bad: 1.7MB/sec. The systems is (in case you hadn't already guessed) a Compaq ProLiant DL380-G2 with 1GB main memory and dual 1266MHz CPUs. I will attach dmesg, my test program, and the output of bioctl -iv ami0 below. I can't imagine the MegaRAID controller is this slow. Any suggestions? can you try running multiple copies of your test app at the same time and see if performance increases? regards, dlg Rob Urban dmesg: --- OpenBSD 3.9 (GENERIC) #617: Thu Mar 2 02:26:48 MST 2006 [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC cpu0: Intel(R) Pentium(R) III CPU family 1266MHz (GenuineIntel 686-class) 1.27 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE3 6,M MX,FXSR,SSE real mem = 1073307648 (1048152K) avail mem = 972660736 (949864K) using 4278 buffers containing 53768192 bytes (52508K) of memory mainbus0 (root) bios0 at mainbus0: AT/286+(00) BIOS, date 12/31/99, BIOS32 rev. 0 @ 0xf pcibios0 at bios0: rev 2.1 @ 0xf/0x2000 pcibios0: PCI BIOS has 9 Interrupt Routing table entries pcibios0: PCI Interrupt Router at 000:15:0 (ServerWorks OSB4 rev 0x00) pcibios0: PCI bus #0 is the last bus bios0: ROM list: 0xc/0x8000 0xc8000/0x2600 0xee000/0x2000! cpu0 at mainbus0 pci0 at mainbus0 bus 0: configuration mode 1 (no bios) pchb0 at pci0 dev 0 function 0 ServerWorks CNB20HE Host rev 0x23 pci1 at pchb0 bus 1 ppb0 at pci1 dev 3 function 0 Intel i960 RP PCI-PCI rev 0x05 pci2 at ppb0 bus 2 vga1 at pci2 dev 0 function 0 ATI Mach64 GV rev 0x7a wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation) wsdisplay0: screen 1-5 added (80x25, vt100 emulation) Intel 80960RP ATU rev 0x05 at pci1 dev 3 function 1 not configured pchb1 at pci0 dev 0 function 1 ServerWorks CNB20HE Host rev 0x01 pchb2 at pci0 dev 0 function 2 ServerWorks CNB20HE Host rev 0x01 pchb3 at pci0 dev 0 function 3 ServerWorks CNB20HE Host rev 0x01 pci3 at pchb3 bus 7 ami0 at pci3 dev 4 function 0 Symbios Logic MegaRAID rev 0x01: irq 3 Dell 518 64b/lhc ami0: FW 351X, BIOS v1.10, 128MB RAM ami0: 2 channels, 0 FC loops, 2 logical drives scsibus0 at ami0: 40 targets sd0 at scsibus0 targ 0 lun 0: AMI, Host drive #00, SCSI2 0/ direct fixed sd0: 208110MB, 208110 cyl, 64 head, 32 sec, 512 bytes/sec, 426209280 sec total sd1 at scsibus0 targ 1 lun 0: AMI, Host drive #01, SCSI2 0/ direct fixed sd1: 286080MB, 286080 cyl, 64 head, 32 sec, 512 bytes/sec, 585891840 sec total scsibus1 at ami0: 16 targets uk0 at scsibus1 targ 15 lun 0: COMPAQ, PROLIANT 4L6I, 1.84 SCSI2 3/processor fixed uk0: unknown device scsibus2 at ami0: 16 targets Compaq PCI Hotplug rev 0x12 at pci3 dev 7 function 0 not configured fxp0 at pci0 dev 2 function 0 Intel 8255x rev 0x08, i82559: irq 5, address 00:08:02:8a:4b:fc inphy0 at fxp0 phy 1: i82555 10/100 PHY, rev. 4 fxp1 at pci0 dev 4 function 0 Intel 8255x rev 0x08, i82559: irq 7, address 00:08:02:8a:4b:fb inphy1 at fxp1 phy 1: i82555 10/100 PHY, rev. 4 Compaq Netelligent ASMC rev 0x00 at pci0 dev 6 function 0 not configured piixpm0 at pci0 dev 15 function 0 ServerWorks OSB4 rev 0x51: SMBus disabled pciide0 at pci0 dev 15 function 1 ServerWorks OSB4 IDE rev 0x00: DMA atapiscsi0 at pciide0 channel 0 drive 0 scsibus3 at atapiscsi0: 2 targets cd0 at scsibus3 targ 0 lun 0: COMPAQ, CD-ROM SN-124, N102 SCSI0 5/ cdrom removable cd0(pciide0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 1 ohci0 at pci0 dev 15 function 2 ServerWorks OSB4/CSB5 USB rev 0x04: irq 11, version 1.0, legacy support usb0 at ohci0: USB revision 1.0 uhub0 at usb0 uhub0: ServerWorks OHCI root hub, rev 1.00/1.00, addr 1 uhub0: 4 ports with 4 removable, self powered isa0 at mainbus0 isadma0 at isa0 pckbc0 at isa0 port 0x60/5 pckbd0 at pckbc0 (kbd slot) pckbc0: using irq 1 for kbd slot wskbd0 at pckbd0: console keyboard,
Re: MegaRAID SCSI 320-2 bad write performance
no battery = no cache. Performance is abysmal without. On Tue, Aug 22, 2006 at 02:17:35AM +0200, Robert Urban wrote: Hi Folks, using a simple test program to write sequential blocks to a file, optionally opening with O_SYNC, I've tested write performance to a MegaRAID logical drive consisting of a RAID-5 set of 4 72GB HP Ultra320 disks and to a RAID-0 drive consisting of a single 300GB HP Ultra320 disk. The controller has 128MB of cache, but I do not have a battery, so cache write policy is write-through. The kernel is the bsd, and not bsd.mp, but it makes no difference. I've tried both. All tests performed with O_SYNC, to avoid bufcache interaction. The performance, at least to my perhaps naive eyes, seems abysmal. I'm getting 1.2MB/sec on the RAID-5 logical drive. It doesn't matter what blocksize I choose, 32k, 64k, 128k, 256k. The chunksize on the logical drive is the default, 64k. On the single HP 300GB Ultra320 RAID-0 drive write performance is also bad: 1.7MB/sec. The systems is (in case you hadn't already guessed) a Compaq ProLiant DL380-G2 with 1GB main memory and dual 1266MHz CPUs. I will attach dmesg, my test program, and the output of bioctl -iv ami0 below. I can't imagine the MegaRAID controller is this slow. Any suggestions? Rob Urban dmesg: --- OpenBSD 3.9 (GENERIC) #617: Thu Mar 2 02:26:48 MST 2006 [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC cpu0: Intel(R) Pentium(R) III CPU family 1266MHz (GenuineIntel 686-class) 1.27 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,M MX,FXSR,SSE real mem = 1073307648 (1048152K) avail mem = 972660736 (949864K) using 4278 buffers containing 53768192 bytes (52508K) of memory mainbus0 (root) bios0 at mainbus0: AT/286+(00) BIOS, date 12/31/99, BIOS32 rev. 0 @ 0xf pcibios0 at bios0: rev 2.1 @ 0xf/0x2000 pcibios0: PCI BIOS has 9 Interrupt Routing table entries pcibios0: PCI Interrupt Router at 000:15:0 (ServerWorks OSB4 rev 0x00) pcibios0: PCI bus #0 is the last bus bios0: ROM list: 0xc/0x8000 0xc8000/0x2600 0xee000/0x2000! cpu0 at mainbus0 pci0 at mainbus0 bus 0: configuration mode 1 (no bios) pchb0 at pci0 dev 0 function 0 ServerWorks CNB20HE Host rev 0x23 pci1 at pchb0 bus 1 ppb0 at pci1 dev 3 function 0 Intel i960 RP PCI-PCI rev 0x05 pci2 at ppb0 bus 2 vga1 at pci2 dev 0 function 0 ATI Mach64 GV rev 0x7a wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation) wsdisplay0: screen 1-5 added (80x25, vt100 emulation) Intel 80960RP ATU rev 0x05 at pci1 dev 3 function 1 not configured pchb1 at pci0 dev 0 function 1 ServerWorks CNB20HE Host rev 0x01 pchb2 at pci0 dev 0 function 2 ServerWorks CNB20HE Host rev 0x01 pchb3 at pci0 dev 0 function 3 ServerWorks CNB20HE Host rev 0x01 pci3 at pchb3 bus 7 ami0 at pci3 dev 4 function 0 Symbios Logic MegaRAID rev 0x01: irq 3 Dell 518 64b/lhc ami0: FW 351X, BIOS v1.10, 128MB RAM ami0: 2 channels, 0 FC loops, 2 logical drives scsibus0 at ami0: 40 targets sd0 at scsibus0 targ 0 lun 0: AMI, Host drive #00, SCSI2 0/direct fixed sd0: 208110MB, 208110 cyl, 64 head, 32 sec, 512 bytes/sec, 426209280 sec total sd1 at scsibus0 targ 1 lun 0: AMI, Host drive #01, SCSI2 0/direct fixed sd1: 286080MB, 286080 cyl, 64 head, 32 sec, 512 bytes/sec, 585891840 sec total scsibus1 at ami0: 16 targets uk0 at scsibus1 targ 15 lun 0: COMPAQ, PROLIANT 4L6I, 1.84 SCSI2 3/processor fixed uk0: unknown device scsibus2 at ami0: 16 targets Compaq PCI Hotplug rev 0x12 at pci3 dev 7 function 0 not configured fxp0 at pci0 dev 2 function 0 Intel 8255x rev 0x08, i82559: irq 5, address 00:08:02:8a:4b:fc inphy0 at fxp0 phy 1: i82555 10/100 PHY, rev. 4 fxp1 at pci0 dev 4 function 0 Intel 8255x rev 0x08, i82559: irq 7, address 00:08:02:8a:4b:fb inphy1 at fxp1 phy 1: i82555 10/100 PHY, rev. 4 Compaq Netelligent ASMC rev 0x00 at pci0 dev 6 function 0 not configured piixpm0 at pci0 dev 15 function 0 ServerWorks OSB4 rev 0x51: SMBus disabled pciide0 at pci0 dev 15 function 1 ServerWorks OSB4 IDE rev 0x00: DMA atapiscsi0 at pciide0 channel 0 drive 0 scsibus3 at atapiscsi0: 2 targets cd0 at scsibus3 targ 0 lun 0: COMPAQ, CD-ROM SN-124, N102 SCSI0 5/cdrom removable cd0(pciide0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 1 ohci0 at pci0 dev 15 function 2 ServerWorks OSB4/CSB5 USB rev 0x04: irq 11, version 1.0, legacy support usb0 at ohci0: USB revision 1.0 uhub0 at usb0 uhub0: ServerWorks OHCI root hub, rev 1.00/1.00, addr 1 uhub0: 4 ports with 4 removable, self powered isa0 at mainbus0 isadma0 at isa0 pckbc0 at isa0 port 0x60/5 pckbd0 at pckbc0 (kbd slot) pckbc0: using irq 1 for kbd slot wskbd0 at pckbd0: console keyboard, using wsdisplay0 pms0 at pckbc0 (aux slot) pckbc0: using irq 12 for aux slot wsmouse0 at pms0 mux 0 pcppi0 at isa0 port 0x61 midi0 at pcppi0: PC speaker spkr0 at pcppi0 npx0 at isa0
