| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 475213 | Computers & Operations Research | 2012 | 6 Pages |
Abstract
A parallel implementation via CUDA of the dynamic programming method for the knapsack problem on NVIDIA GPU is presented. A GTX 260 card with 192 cores (1.4 GHz) is used for computational tests and processing times obtained with the parallel code are compared to the sequential one on a CPU with an Intel Xeon 3.0 GHz. The results show a speedup factor of 26 for large size problems. Furthermore, in order to limit the communication between the CPU and the GPU, a compression technique is presented which decreases significantly the memory occupancy.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science (General)
Authors
V. Boyer, D. El Baz, M. Elkihel,