# Question about Memory Workload Analysis (keyword)

**URL:** <https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373>\
**Category:** Nsight Compute\
**Created:** [June 21, 2021, 4:26am UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373 "2021-06-21T04:26:05Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![sakaia](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@sakaia](https://forums.developer.nvidia.com/u/sakaia)\
**Post date:** [June 21, 2021, 4:26am UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373/1 "2021-06-21T04:26:05Z")

</div>

Question about Memory Workload Analysis.

does there exist any documentation about Memory Workload Analysis?

Especially I want to know the keywords in map. It seems very helpful but I cannot found documentation in [https://docs.nvidia.com/nsight-compute/NsightCompute/index.html](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html)

I am attached the samples on Nsight Compute with running tf32TensorCoreGemm in cuda-samples.

[ncu\_cudasamples\_tf32.pdf](https://forums.developer.nvidia.com/uploads/short-url/xhGJsZPcwQufpK1hJWgAfo53kz7.pdf) (1.2 MB)

---

<div class="post-metadata">

**Author:** ![felix\_dt](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/felix_dt/32/14043_2.png) [@felix\_dt](https://forums.developer.nvidia.com/u/felix_dt)\
**Post date:** [June 21, 2021, 5:34am UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373/2 "2021-06-21T05:34:57Z")

</div>

You can find information about the memory workload charts and tables [here](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html#memory-tables). Further info about the used terms can be found in the [metrics guide](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html#metrics-decoder).

---

<div class="post-metadata">

**Author:** ![sakaia](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@sakaia](https://forums.developer.nvidia.com/u/sakaia)\
**Post date:** [June 21, 2021, 5:40am UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373/3 "2021-06-21T05:40:28Z")

</div>

Thank you for your reply.

In summary, The user should check both documents, i.e. Nsight Compute and Kernel Profiling Guide?

---

<div class="post-metadata">

**Author:** ![felix\_dt](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/felix_dt/32/14043_2.png) [@felix\_dt](https://forums.developer.nvidia.com/u/felix_dt)\
**Post date:** [June 21, 2021, 11:44am UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373/4 "2021-06-21T11:44:11Z")

</div>

That’s right. The Nsight Compute document primarily explains the workflows and elements of the user interface, while the Kernel Profiling Guide documents the metric terminology and meaning itself and provides more insight on interpreting the data.

---

<div class="post-metadata">

**Author:** ![nadeemm](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/nadeemm/32/367020_2.png) [@nadeemm](https://forums.developer.nvidia.com/u/nadeemm)\
**Post date:** [October 12, 2021, 6:07pm UTC](https://forums.developer.nvidia.com/t/question-about-memory-workload-analysis-keyword/181373/5 "2021-10-12T18:07:03Z")

</div>

Closed on October 12, 2021, 6:07pm UTC