Dear =?ISO-8859-15?Q?Stefan_/*St0fF*/_H=FCbner?=, In message <4D361F26.3060507@xxxxxxxxxxxxxxxxxx> you wrote: > > [in German:] Schätzelein, Dein Problem sind die Platten, nicht der > Controller. Irrtum. > [in English:] Dude, the disks are your bottleneck. Wrong. Testing the same workload with soft RAID versus the h/w RAID solution gives a _significant_ performance difference. I happen to know which benchmarks 3ware (and other RAID controller manufacturers) are optimizing their firmware for - IOPS is not even mentioned there. > On a 4-disk RAID0 software RAID can only outspeed this 3ware Controller > with a really really fast processor. The limiting factor is the disk's > access time. If SSDs are too expensive, then your actual performance is > the max you'll get (maybe to replace the HWRAID controller might give a > little speed-up, but not very much). >From some tests done before I expect to see a speed increase of >10. Hey, even a single disk is performing better under this work load. And fast processor? Yes, I have it, but what for? It spends most of it's time (>90%, usually more) in iowait. Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@xxxxxxx "To IBM, 'open' means there is a modicum of interoperability among some of their equipment." - Harv Masterson -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
