Subscribe to this blog
Briefcase
Print
We got something good in the mail last week!
Some quick observations:
The build quality is outstanding. Nothing cheap at all about this card. The engineering that has gone into this shows in every way.
It is made up of modules that are screwed down, I can see where they really thought this through so each rev of the card doesn’t require all new PCB’s to be manufactured.
It does require an external source of power via 4 pin Molex or SATA power connector period. Make sure your server has one available, even though these are sold by HP not all HP servers have the required connectors.
PCIe expander bays are few and far between. The issue is most of these are used to expand desktops, laptops or used in non critical applications mostly AV or render farms.
http://www.magma.com/products/pciexpress/expressbox4-1u/index.html
This is a nice chassis but they are currently being retooled and won’t be available for a month or so. It is the only 1U and it has redundant power.
It exposes two drives to the OS per card. We will initially configure them two per machine in a RAID 10 array for redundancy.
More to come!
Wes
Posted by Steve Jones on 9 June 2009
Interesting, I'd love to know more about the before/after config as well as performance.
Posted by Wesley Brown on 9 June 2009
We are testing first on a machine that just came out of production this week. So I'll have some apples to apples kind of numbers.
I'm documenting everything as much as I am able to and setting up to capture as much data as possible :)
Posted by RichardB on 15 June 2009
I have 3 in a dev box at the moment - absolutely awesome.
Truly insane throughput, and the latencies are teeny tiny.
Really impressive performance booster.
Only issue is that it will be a true shared nothing architecture, so will need to mirror rather than cluster, but I am hoping (and testing...) that the throughput will be handled by these.
If I do get them into production will let you all know, but in the meantime, if you can get your hands on some of these just do it!
Rich
Posted by Wesley Brown on 15 June 2009
I'm almost done with my first synthetic benchmark pass using SQLIO so I'll have an update in a day or two. Some interesting stuff with different queue depths and IO block sizes compared with the disk setup also on the server.