http://www.linux-mag.com/id/7393/ Open Source, Open Standards Sat, 05 Oct 2013 13:48:18 +0000 hourly 1 http://wordpress.org/?v=3.1 http://www.linux-mag.com/id/7393/#comment-981023 Answers Tue, 04 Jun 2013 08:48:20 +0000 http://www.linux-mag.com/id/7393/#comment-981023 Now lets turn back the clock a bit. I dug up some old Stream numbers for dual core Opterons. A four core Opteron system (two dual core processors) was able to deliver 12 GBytes/sec or 3 GBytes/sec per core. If you <a href="http://eco.cueb.edu.cn/contents/page/" rel="nofollow">Cueb Answers</a> were to run a memory bound MPI application on 24 cores, what would you want to use? A single 24 core SMP workstation (total memory bandwidth of 41 GBytes/sec) or six dual socket, dual core nodes (total combined memory bandwidth of 72 GBytes/sec). Now lets turn back the clock a bit. I dug up some old Stream numbers for dual core Opterons. A four core Opteron system (two dual core processors) was able to deliver 12 GBytes/sec or 3 GBytes/sec per core. If you Cueb Answers were to run a memory bound MPI application on 24 cores, what would you want to use? A single 24 core SMP workstation (total memory bandwidth of 41 GBytes/sec) or six dual socket, dual core nodes (total combined memory bandwidth of 72 GBytes/sec).

]]> http://www.linux-mag.com/id/7393/#comment-6597 karlarao Wed, 30 Nov -0001 00:00:00 +0000 http://www.linux-mag.com/id/7393/#comment-6597 <p>What was the tool used to measure the stream bandwidth of 41 GBytes/sec? </p> What was the tool used to measure the stream bandwidth of 41 GBytes/sec?

]]>