Example optimisation using GPGPUs by ICHEC

857 views
812 views

Published on

Example optimisation using GPGPUs by ICHEC of a Monte Carlo based simulation for a real client.

Published in: Technology, Business
1 Comment
0 Likes
Statistics
Notes
  • Replaced the earlier one as it wasn't clear that the first item was a reference that we based on performance improvements on. Additionally our first port to C++ did show some significant gains to Java before it was even multithreaded prior to its CUDA GPGPU optimisation.
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

No Downloads
Views
Total views
857
On SlideShare
0
From Embeds
0
Number of Embeds
5
Actions
Shares
0
Downloads
0
Comments
1
Likes
0
Embeds 0
No embeds

No notes for slide

Example optimisation using GPGPUs by ICHEC

  1. 1. Application.version Number.of.Simulations 100 101 102 103 104 105 106 100 101 102 103 104 105 106 100 101 102 103 104 105 106 Test 1 Test 4 Test 7 Java 1 Thread C++ 1 Thread C++ 12 Threads x1 GPU x2 GPUs Test 2 Test 5 Test 8 Java 1 Thread C++ 1 Thread C++ 12 Threads x1 GPU x2 GPUs Test 3 Test 6 Java 1 Thread C++ 1 Thread C++ 12 Threads x1 GPU x2 GPUs Application.ver Java 1 Thre C++ 1 Thre C++ 12 Thr x1 GPU x2 GPUs Optimisation of Monte Carlo Simulations by ICHEC Performance improvements are based on comparison with original Java code. x1.9 Number of simulations completed in 500ms LOG LOG LOG x2 Code to be optimised Test 3 Test 6 Application.version Java 1 Thread C++ 1 Thread C++ 12 Threads x1 GPU x2 GPUs 1 Thread, Java Reference 1 Thread, C++ 12 Threads, C++ x1 GPU x2 GPUs x1.8 x2.1 x2 x1.9 x2x2 x12 x12 x11 x12 x12 x12 x12x12 x91 x92 x76 x100 x93 x82 x311x91 x182 x184 x149 x193 x176 x160 x599x174

×