Development Tools Nsight Compute
Topic | Replies | Views | Activity | |
---|---|---|---|---|
NVIDIA® Nsight™ Compute 2023.1 is now available
NVIDIA® Nsight™ Compute 2023.1 is now available for download in the NVIDIA Registered Developer Program. Nsight™ Compute 2023.1.0 supports the CUDA Toolkit 12.1, OpitX 7.7 and a new application range replay mode. NVI… |
![]() |
4 | 682 | February 28, 2023 |
NVIDIA® Nsight™ Compute 2022.4 is now available
NVIDIA® Nsight™ Compute 2022.4 is now available for download in the NVIDIA Registered Developer Program. Nsight™ Compute 2022.4.0 supports the CUDA Toolkit 12.0, and profiling on WSL2. NVIDIA® Nsight™ Compute 2022.4 … |
![]() |
4 | 926 | December 9, 2022 |
NVIDIA® Nsight™ Compute 2022.2 is now available
NVIDIA® Nsight™ Compute 2022.2 is now available for download in the NVIDIA Registered Developer Program. Nsight™ Compute 2022.2.0 supports the CUDA Toolkit 11.7, Nsight Compute (GUI) is now available on aarch64 SBSA, a… |
![]() |
4 | 1802 | May 11, 2022 |
NVIDIA® Nsight™ Compute 2021.1 is now available
NVIDIA® Nsight™ Compute 2021.1.0 is now available for download in the NVIDIA Registered Developer Program. Version 2021.1.0 supports the CUDA Toolkit 11.3, Optix 7 API, and NVIDIA’s latest Ampere architecture GPUs. The… |
![]() |
2 | 1736 | April 15, 2021 |
NVIDIA® Nsight™ Compute 2020.3 is now available
NVIDIA® Nsight™ Compute 2020.3.0 is now available for download in the NVIDIA Registered Developer Program. Version 2020.3 supports the CUDA Toolkit 11.2 and NVIDIA’s latest Ampere architecture GPUs. Along with many w… |
![]() |
4 | 2134 | December 15, 2020 |
About the Nsight Compute category
|
![]() |
0 | 1651 | February 1, 2020 |
Different achieved values in Roofline
|
![]() |
0 | 14 | June 5, 2023 |
Kernel orders when using different metrics
|
![]() ![]() |
6 | 436 | June 2, 2023 |
Profiling one application having two concurent kernels
|
![]() ![]() |
2 | 47 | June 2, 2023 |
Cannot profile kernel from CUDA samples
|
![]() ![]() |
6 | 128 | May 31, 2023 |
How can I get the tile size of gemm for cudnn kernel name?
|
![]() ![]() |
1 | 83 | May 24, 2023 |
Profiling using Nsight compute CLI python script with Cupy
|
![]() ![]() |
1 | 123 | May 23, 2023 |
NSight Profiling Crashes with error code (9)
|
![]() ![]() ![]() ![]() |
4 | 780 | May 18, 2023 |
Error failed to profile kernel
|
![]() ![]() |
2 | 143 | May 18, 2023 |
Calculation of Memory Bound nature vs Roofline numbers
|
![]() ![]() |
3 | 455 | May 18, 2023 |
Is it possible to get detailed info about GPU memory specifically shared memory?
|
![]() ![]() |
3 | 147 | May 17, 2023 |
Unable to profile with NCU -- WARNING: No Kernels were profiled
|
![]() ![]() ![]() ![]() |
3 | 706 | May 15, 2023 |
How to profile overall SM utilization of the program by Nsight Compute?
|
![]() ![]() |
7 | 776 | May 11, 2023 |
Did fp16 roofline support ever make it into ncu?
|
![]() ![]() |
1 | 202 | May 11, 2023 |
Shared memory: LDGSTS Async copy understanding problem
|
![]() ![]() |
1 | 180 | May 11, 2023 |
There is a big difference between nsc and theoretical Arithmetic Intensity. Is it resonable?
|
![]() ![]() |
5 | 185 | May 11, 2023 |
Same kernel 3x slower on CUDA than on OpenCL
|
![]() ![]() ![]() ![]() |
7 | 247 | May 5, 2023 |
Warp stalls are concentrated on "LDL" instructions
|
![]() ![]() ![]() |
3 | 296 | April 27, 2023 |
Ncu to produce similar output as nvprof
|
![]() ![]() |
1 | 192 | April 27, 2023 |
L2 cache rate profiled in nsight compute is confused
|
![]() ![]() |
3 | 553 | April 26, 2023 |
What does the metric "pcie throughput" mean specifically?
|
![]() ![]() ![]() |
6 | 509 | April 26, 2023 |
Long/Short Scoreboard Stall
|
![]() ![]() |
1 | 360 | April 24, 2023 |
Reading performance counters vs. disabling frequency throttling
|
![]() ![]() ![]() |
5 | 371 | April 24, 2023 |
Using Nsight Compute (ncu) alongside srun
|
![]() ![]() ![]() ![]() ![]() |
6 | 1509 | April 24, 2023 |
How can I measure kernel launch overhead using ncu
|
![]() ![]() ![]() ![]() |
7 | 287 | May 4, 2023 |