HP is running: 151012
Sunfire: 151010
Supermicro: 151014
Supermicro specs:
_______________________________________________________________
Item Number | Item Type
| # of Units |
CSE-216BAC-R920LPB | Chassis
|
LSI 9300-8i | Disk Controller
| 3.00 |
X10DRi-T4+ | Motherboard
| 1.00 |
Micron/Crucial CT16G4RFD4213 | 16 GB Ram |
12.00 |
Samsung M393A4K40BB0-CPB | 32 GB Ram | 8.00
|
Seagate ST2000NX0243 | 2TB Drive
| 12.00 |
Intel Xeon E5-2637 v3 Quad-core | 3.5 GHz CPU |
2.00 |
--------------------------------------------------------------------------------------------------------
The benchmarks shouldn't be CPU bound in the case of the Supermicro. It has
twice the # of processors of the HP but the HP has 8 cores total in the one
processor (Xeon E5-2640 @ 2ghz). Postgres prefers higher CPU frequency rather
than number of cores. The Sunfire has 48 drives and so writes should be spread
fairly evenly, reducing writes per drive. Is there any other information I can
provide?
Josh
-----Original Message-----
From: qutic development [mailto:mailingli...@qutic.com]
Sent: Thursday, June 23, 2016 6:11 AM
To: Josh Barton <josh.bar...@usurf.usu.edu>
Cc: omnios-discuss <omnios-discuss@lists.omniti.com>
Subject: Re: [OmniOS-discuss] ZPOOL disk performance on Supermicro with LSI
9300-8i
> Am 23.06.2016 um 01:06 schrieb Josh Barton <josh.bar...@usurf.usu.edu>:
>
> Any ideas why the HP is so much faster? It just has one Smart Array
> Controller which I didn’t think would be faster than JBOD
Could you please provide a few more information about the servers? CPU speed,
cores, RAM, etc?
OmniOS tuned or all raw installations? All the same OmniOS version?
- Stefan
_______________________________________________
OmniOS-discuss mailing list
OmniOS-discuss@lists.omniti.com
http://lists.omniti.com/mailman/listinfo/omnios-discuss
