|October 19, 2012, 10:27||
CPU selection (E5-2650) for CFD simulation
Join Date: Mar 2009
Location: Busan, Korea
Posts: 14Rep Power: 7
I'm trying to select a workstation for my CFD simulation. Someone gave me a list of HP Z820 workstation, you can see below, but the price is soooo expensive.
HP Z820 Workstation
HP Single Unit Packaging
HP Z820 1125W 90 Efficient Chassis
Microsoft Windows 7 Professional Edition 64bit OS UK
Windows 7 64 Bit Factory Image Recovery
Intel Xeon E5-2687W 3.10Ghz 20MB 1600 8C 1st CPU
Intel Xeon E5-2687W 3.10Ghz 20MB 1600 8C 2nd CPU
32GB DDR3-1600 (8x4GB) 2CPU Registered RAM
NVIDIA Quadro 2000 1GB Graphics
1TB 7200 RPM SATA 1st Hard Drive
HP USB Standard Keyboard
HP USB Optical Scroll Mouse
16X SuperMulti DVDRW SATA 1st ODD
HP 3/3/3 Warranty EURO
HP Liquid Cooling Solution
HP Z820 Localization Kit
HP Simple Asset Tagging Service
HP YQ205AV Basic UNIT Asset Tag (SN&MAC)
HP YQ206AV Basic BOX Asset Tag (SN&MAC)
HP CPQ LA2206xc 21.5-In LED LCD
HP 3 year Next Business Day Onsite plus Defective Media Retention Workstation
Since the high price, I want to change to CPU to E5-2650. but the frequency of E5-2650 is just 2.0GHz, I'm not sure if this is fast enough for CFD simulation. However, in terms of cpubenchmark, http://www.cpubenchmark.net/high_end_cpus.html, this one is also OK. So, could you give me some suggestion about this workstation?
Should I change some components like RAM or Graphic card?
Thanks a lot!
|October 22, 2012, 12:03||
Join Date: Mar 2009
Location: Austin, TX
Posts: 126Rep Power: 8
If you are using an unstructured mesh, then memory bandwidth is usually more important than CPU clockspeed.
|October 23, 2012, 16:09||
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 430Rep Power: 8
What software will you be using?
You can probably ditch the ECC memory to save money, just go with desktop memory.
|Thread||Thread Starter||Forum||Replies||Last Post|
|FSI TWO-WAY SIMULATION||Smagmon||CFX||1||March 6, 2009 13:24|
|Continuous vs interrupted simulation||sega||OpenFOAM Running, Solving & CFD||4||November 3, 2008 14:29|
|Overflow problem in steady simulation||ReeKo||CFX||11||October 8, 2008 17:57|
|Fire simulation using FDS from NIST||Jens||Main CFD Forum||1||January 22, 2004 01:53|
|3-D Contaminant Dispersal Simulation||Apple L S Chan||Main CFD Forum||1||December 23, 1998 10:06|