Currently running CUDA application against two Kepler K20c GPU devices, and recently upgraded Cuda toolkint, and Linux OS. I was hoping to see similiar throughput performance after upgrading the Linux OS (Centos 5.5 → 7.4), and migrating from Cuda 5.5 to 9.1. However, surprisingly the the GPU application is running 1.5x slower under Cuda 9.1?! To me, it’s not obvious what could be causing the slow down. Has anyone run into a similiar issue after upgrading the Cuda toolkit?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| K20:Lower performance with a higher Driver Verion ? | 1 | 440 | March 7, 2018 | |
| GPU is slower than CPU on my wsl | 2 | 1285 | September 9, 2020 | |
| Upgrading Gfx Card and Cuda (to 8.0 RC) Causes Slowdown | 3 | 686 | October 24, 2016 | |
| Much worse performance after updating cuda toolkit from 10.2 to 11.4 | 3 | 587 | January 20, 2022 | |
| cuSparse performance issue after CUDA Toolkit 9.0: cuSparse<t>csrsv_analysis() 20 times slower | 0 | 530 | February 27, 2018 | |
| Performance drop when changing Toolkit/SDK/driver from 4.2 ---> 5.0 | 14 | 2802 | November 28, 2013 | |
| CUDA 8.0 RC (cuda_8.0.27) is much faster than CUDA 8.0 (cuda_8.0.44) on Pascal | 4 | 2125 | October 20, 2016 | |
| Comparing Cuda 9.2 to Cuda 10.2 | 0 | 394 | May 27, 2020 | |
| Possible issues with CUDA 11.0 on Linux or with an obsolete graphic card | 3 | 1355 | June 29, 2020 | |
| GTX 780 slower than GTX 480? | 4 | 997 | September 29, 2014 |