Is GPU sharing (time-slicing, MPS, MIG, etc.) supported in nvidia nvenc and nvdec?
Also, if sharing is possible, can resources be set appropriately? For example, in MPS, memory limits can be set through CUDA_MPS_PINNED_DEVICE_MEM_LIMIT.
anyeon
1
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| NVDEC with MPS | 5 | 769 | October 12, 2021 | |
| NVEnc licensing | 0 | 470 | November 22, 2021 | |
| Geforce 940MX | 0 | 1064 | August 15, 2016 | |
| Reading gpu memory from cpu | 3 | 19 | June 16, 2025 | |
| Virtualization of CUDA programs | 1 | 651 | October 15, 2018 | |
| NVENC compatibility with NVS | 5 | 1851 | October 12, 2021 | |
| How to sharing GPU Memory between different GPU instance in MIG? | 0 | 746 | March 25, 2022 | |
| Multiple tensorflow sessions running on the same GPU in parallel (with/without MPS) | 0 | 688 | October 7, 2020 | |
| Hyper-Q for sharing GPUs | 1 | 635 | April 7, 2017 | |
| How can I get NVENC capabilities from a C# program ? | 2 | 1944 | November 4, 2015 |