Bad Performance - 4.12E Firmware on 33x0 arrays

Any thoughts?! We just bought two StorEdge 3320, dual controller, 512MB cache each, to replace some 3rd party Ultra160 arrays. These things write 3x slower then the arrays we are trying to replace. Is this possible? So far from Sun Solutions, thier recommendation was to remove the secondary controller since cache mirroring affects a performance. When I did that it did help write speeds but not ready to believe these things are slower then the ones we are replacing with half the throughput and half the caching memory. Just trying to get some insight to maybe help me improve it if it can.

-Paul

[619 byte] By [] at [2007-11-25 23:03:45]

«« Attaching a StorEdge A1000 to a different machine
»» 3510 network interfaces and # of direct attached hosts

# 1

Have you shared the array between two hosts?

Also, there is a known issue with that firmware where performance is degraded by around 10 to 15% over previous versions. It is supposed to make your data safer..

Have you noticed that it takes ages to remove losts of files from the array?

The problem with these things is that are very configurable and some of the options may be incorrectly set. If Sun installed them, the EIS methodology may have corrected some of the issues.

Stephen

stephen2602 at 2007-7-5 17:55:33 >

Storage General Discussion

Java programming resources: FAQs, tutorials, compiler and browser download sites, documentation, books lists, IDEs, etc.

Bad Performance - 4.12E Firmware on 33x0 arrays

Storage Forums New Topic

Storage Forums Hot Topic

Storage Forums

All Classified