looking at the specs between two cards, the SATA card(s) seem to rate ~100-150MB/s on each channel (if I'm reading right), with both the 3Ware and ICP cards having 4 individual channels ... looking at the SCSI cards, they are rated at 320MB/s, but that is total for the SCSI bus itself, right? So, if I have three drives on a SCSI bus, each 'maxing out evenly', I'd be cap'd at about the same 100MB/s per drive, no? In fact, looking at the SATA 2.x specs, each chanell there is rated at 300MB/s, which, again, if I could 'max out evenly', could seriously blow away the SCSI bus itself ... *If* I'm reading this right ... ? ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
On Jun 26, 2005, at 22:34, Marc G. Fournier wrote:> In fact, looking at the SATA 2.x specs, each chanell there is rated at > 300MB/s, which, again, if I could 'max out evenly', could seriously > blow away the SCSI bus itself ... > > *If* I'm reading this right ... ?Bus speed does not equal drive speed. And while yes, SATA is now approaching SCSI in bus speed, it doesn't mean that SCSI isn't standing still. A 640 MB/s bus was standardized in 2003 (though SATA is planned to go to 600MB/s in 2007): http://en.wikipedia.org/wiki/SCSI#Ultra-640 This has been hashed out many times over the years in comp.periph.scsi (it used to be IDE versus SCSI, not it's (S)ATA versus SCSI). A search through the archives would probably return many results. The SCSI FAQ [1] would also probably be useful. Specifically the questions: QUESTION: What are the pros and cons regarding SCSI vs IDE/ATA ? http://home.comcast.net/~SCSIguy/SCSI_FAQ/scsifaq.html#_Hlk407004722 QUESTION:Should I spend the extra money on SCSI or just get IDE? http://home.comcast.net/~SCSIguy/SCSI_FAQ/scsifaq.html#_Hlk407091459 QUESTION:Why do SCSI disks cost so much more than IDE/ATA disks? http://home.comcast.net/~SCSIguy/SCSI_FAQ/scsifaq.html#_SCSI_Cost002 [1] http://www.scsifaq.org/
On Sun, Jun 26, 2005 at 11:34:22PM -0300, Marc G. Fournier wrote:> > looking at the specs between two cards, the SATA card(s) seem to rate > ~100-150MB/s on each channel (if I'm reading right), with both the 3Ware > and ICP cards having 4 individual channels ... looking at the SCSI cards, > they are rated at 320MB/s, but that is total for the SCSI bus itself, > right?It is dependant of the card too. You can get a 3 channel Ultra320 card for 3 drive RAID 5...> So, if I have three drives on a SCSI bus, each 'maxing out evenly', I'd > be cap'd at about the same 100MB/s per drive, no? > > In fact, looking at the SATA 2.x specs, each chanell there is rated at > 300MB/s, which, again, if I could 'max out evenly', could seriously blow > away the SCSI bus itself ...That is a theory...> *If* I'm reading this right ... ?And now from my practice: 1) You CAN'T get maximal throuput from 150/160 MB/s bus - the is no such drives today at the market. 2) Avg. seek and access time of 7200 SATA drives are FAR BELOW such of old Seagate Cheetah/Maxtor Athlas 10K. So is a situation with a WD Raptor 10K SATA too - they are SLOWER! (I'm not shure why, but from my test: 2xPIII-1.4/1G RAM/2x36G Cheeta 10K in RAID 1 (AMR Express; One U160 channel) ws. 2xXeon 3.0/2G RAM/2x72G Raptor in RAID 1 (Intel SRCS16; Separate SATA-150 channel for each drive) Ports three checkout is about 20% faster on the first comp. As an result: If you need to have FAST disk subsystem - buy SCSI. If you need large storage or you don't want to pay to much money - buy SATA. Best Ragards, Serg N. Voronkov, Sibitex JSC.
Marc G. Fournier <scrappy@hub.org> wrote:> looking at the specs between two cards, the SATA card(s) seem to rate > ~100-150MB/s on each channel (if I'm reading right), with both the > 3Ware and ICP cards having 4 individual channels ... looking at the > SCSI cards, they are rated at 320MB/s, but that is total for the SCSI > bus itself, right? > > So, if I have three drives on a SCSI bus, each 'maxing out evenly', > I'd be cap'd at about the same 100MB/s per drive, no? > > In fact, looking at the SATA 2.x specs, each chanell there is rated at > 300MB/s, which, again, if I could 'max out evenly', could seriously > blow away the SCSI bus itself ... > > *If* I'm reading this right ... ?For the last 6 month i really think that if you don't need something high-end scsi then you should go for SATA. There are test on sites such as Tom's hardware guide and ixbt.com. They show then on sequrncial read there is no difference between scsi and sata. Acatuallty, modern hdds use the same mechanics for sata and scsi versions of them. The brains (electronics) on the hdds are different of course. However, when it comes to random read/writes scsi wins because of command queueing. This was an issue until recently, Recently SATA with NCQ became widly available. Test show that some of those SATA disks with NCW ***WIN*** over scsi 320. The test envolve artificialy random read/write tests as well as real application benchmarking. I din't rememeber where excatly i saw the tests on those site, but you could search. So, my opinion, workstation never needs SCSI and every server MUST be on mirror or RAID5 and there you should use SATA with NCQ drivers unless, your applicaton is really weird and needs something extremely speedy. Then, however, you could go for RAID 0+1 and get perfomance that SCSI will never get you. -- Regards, Artem Kuchin IT Legion Ltd. Moscow, Russia www.itlegion.ru matrix@itlegion.ru +7 095 232-0338
Marc G. Fournier wrote:> > looking at the specs between two cards, the SATA card(s) seem to rate > ~100-150MB/s on each channel (if I'm reading right), with both the 3Ware > and ICP cards having 4 individual channels ... looking at the SCSI > cards, they are rated at 320MB/s, but that is total for the SCSI bus > itself, right?[snip] As is often the case, these kind of comparisons are only working against one side of the equation: "What happens when the system is operational" Now, for your average desktop system, the converse: "What happens when things break" becomes negligable. Slap in new drive, slap in DVD/CD install media, Do Stuff[tm], make working system. In the Non-Stop (probably a trademark of Tandem) server-class world, things work somewhat differently. <flamebait>This is why we run postgresql instead of mysql</flamebait>. This is also the reason we run U320 SCSI enclosures at twice the cost of equivalent SATA systems. The extra cash is well spent when things go wrong. Absolute performance (assuming it can be measured correctly for the specific environment you're in, as opposed to marketing figures) counts for less than 50% of the whole equation when it comes to the purchase decision. -aDe