Hi,
I was wondering if there’s a way to get the latency of shared memory and registers. Specifically for the H100.
Thanks!
Hi,
I was wondering if there’s a way to get the latency of shared memory and registers. Specifically for the H100.
Thanks!
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Why register per thread in nsight compute different from nvcc --ptxas-options=-v? | 6 | 844 | January 19, 2023 | |
| [Fermi] Number of registers | 36 | 20498 | September 15, 2010 | |
| latency of shared memory of Tesla C1060 | 0 | 1264 | December 8, 2009 | |
| Does the Roofline Model's L1 Cache Bandwidth Include Shared Memory? | 4 | 83 | October 20, 2025 | |
| Determining Registers Per Work-Item and Shared Memory Per Work-group | 2 | 3811 | November 18, 2011 | |
| Accessing Frequency of different types of data stored in HBM integrated with Nvidia's Processors | 2 | 223 | June 24, 2025 | |
| global memory latency | 6 | 6102 | December 24, 2008 | |
| Metric references and description | 7 | 5176 | March 2, 2024 | |
| what is the mean of `gpu__compute_memory_access_throughput` | 4 | 1080 | August 22, 2019 | |
| How to know my kernel if Pipeline parallel by nsight compute | 6 | 1016 | April 18, 2023 |