Am 30.05.2014 um 19:34 schrieb John R Pierce pierce@hogranch.com:
On 5/30/2014 5:28 AM, Andrew Holway wrote:
As I think about it, the control node for the UV 2000 looks an*awful* lot like a Penguin....
SuperMicro gear is only as good as the server integration company selling it.
bingo. I think too many people buy 'whitebox' supermicro stuff direct and self-integrate, then are surprised when there are issues. Integration needs to include testing. All that integration and testing is why brands like HP are more expensive, you can usually assume its going to work.
True. The thing I hate about HP is that their SSD offerings are IMO a joke.
Not only are they several times as expensive as an equivalent Intel SSD (even taking into account that we don’t pay list-price) but in addition, they perform only half as well (in terms of IOP/s).
I suspect it’s because HP does not include a super cap and thus their SSDs don’t do write-caching (which the Intel does).
Also, due to the fact that they don’t offer a SAS-Controller that does JBOD, you have to setup each drive individually as a RAID0 - which is totally stupid, once you run something like FreeBSD where HPACUCLI is not available. Each failed drive necessitates a reboot then.
I could of course buy an LSI JBOD controller (which would also allow me to buy Intel SSDs) - but what’s the point of buying a HP server then?
IMO, HP does not want you to actually make good use of current-generation enterprise-SSDs - they’d prefer you buy a couple of dozens of P2000 arrays instead…