7200-hardware-picture1
Uncategorized — By Chris Evans on April 28, 2009 at 9:59 AM
(No Ratings Yet)
Loading ...
Tweet This
Digg This
Save to delicious
Stumble it
RSS Feed
(No Ratings Yet) Loading ...-
Virtualisation: Home Lab Update
March 9, 2010 07:03
-
Enterprise Computing: HP Blades Tech Day 2
March 5, 2010 06:03
-
Enterprise Computing: Sun/Oracle Kicks Hitachi To The Kerb
March 2, 2010 22:03
-
Enterprise Computing: HP Blades Tech Day 1
March 2, 2010 11:03
-
Enterprise Computing: HP Blades Tech Day 0
February 24, 2010 22:02
-
Visiotech: Strangely that is exactly what I wrote back in december 14 on Storage Monkeys. http://www.storagemonkeys.com/index....
-
GMAN: I guarantee it will NOT be Netapp. Has anybody out there really run Sun OS with Oracle on Netapp? Well, it's quite a fra...
-
michael p: As a follow up, version 1.6.8 of Drobo Dashboard has been released and it seems to work fine in my configuration that wa...
-
michael p: Hi Chris, It has been a frustrating start, but in the end it is working. Yes, I am using the iSCSI initiator that com...
-
Han Solo: Anyone think Pillar maybe?! Hello!? It is Larry's other company after all....
- @alextangent OK, that clarifies the random I/O comment - thanks
- @alextangent OK, makes sense. So only impact would be if I/O so great, NVRAM gets flushed continuously.
- @alextangent <-- the same set of pages on the host (say 4K/8K). I can't see that being efficient as WAFL more suited to seq write.
- @alextangent I understand the architecture (done all the training), what I'm not sure is the impact on WAFL/NVRAM of high I/O to -->
- @Storagezilla @storagebod Let's do a trip to Cork! Get the undertakers to collect us and organise a massive fry-up brekkie!
- @alextangent I mean if I write to disk, am I required to destage an entire raid stripe to write a single 4K block or just that block?
- @alextangent OK, simple question. Can Data ONTAP accept a write and commit to disk a *single* 4K random block asynchronously?
- @Storagezilla The fun of camping out at LHR.
- @Storagezilla Don't tell me you're in London tomorrow....
- @alextangent depends on whether your underlying architecture is designed to handle full random I/O or not. If not, SAN isn't best choice
