-
Notifications
You must be signed in to change notification settings - Fork 371
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add benchmark results to docs #3244
Add benchmark results to docs #3244
Conversation
Co-authored-by: Dennis Terhorst <[email protected]>
Co-authored-by: Melissa Lober <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for these updates.
|
||
|
||
NEST performance is continuously monitored and improved across various network sizes. | ||
Here we show benchmarking results for NEST version 3.8 on Jureca-DC. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
replace: "for this version of NEST on JURECA-DC [x]."
Jureca is discribed in the large scale instrument Journal of Juelich. Please include the reference in the usual citation format. Here is the paper., it also has a doi: https://jlsrf.org/index.php/lsf/article/view/182/pdf
Add a sentence: "The benchmarking framework and the structure of the graphs is described in [Y]."
Give for [Y] the reference to the beNNch paper in the usual citation format:
Albers J, Pronold J, Kurth AC, Vennemo SB, Haghighi Mood K, Patronis A, Terhorst D, Jordan J, Kunkel S, Tetzlaff T, Diesmann M and Senk J (2022) A Modular Workflow for Performance Benchmarking of Neuronal Network Simulations. Front. Neuroinform. 16:837549. doi: 10.3389/fninf.2022.837549
:class: sd-align-minor-center | ||
|
||
|
||
* The model has ~80 000 neurons and ~300 million synapses |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add: ", minimal delay 0.1ms"
add new bullet item:
- 2 MPI processes per node, 64 threads per MPI process
|
||
* The model has ~80 000 neurons and ~300 million synapses | ||
* Increasing number of computing resources decrease simulation time | ||
* The model runs faster than real time |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add reference [Z] at end of sentence so that the concept of "real time" can be looked up and there is a link for further information: Here is the reference:
Kurth, A. C. ; Senk, J.; Terhorst, D.; Finnerty, J.; Diesmann, M. (2022) Sub-realtime simulation of a neuronal network of natural density. Neuromorphic computing and engineering 2(2), 021001 (2022)
you can add doi link
:columns: 10 | ||
:class: sd-align-minor-center | ||
|
||
* The model has ~4.1 million neurons and ~24 billion synapses |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add: ", minimal delay 0.1ms"
add new bullet item:
2 MPI processes per node, 64 threads per MPI process
:class: sd-align-minor-center | ||
|
||
|
||
* The size of network scales proportionally with the computational resources used |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add: ", minimal delay 1.5ms"
add new bullet item:
2 MPI processes per node, 64 threads per MPI process
|
||
|
||
* The model has ~80 000 neurons and ~300 million synapses | ||
* Increasing number of computing resources decrease simulation time |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add bullet item:
- data averaged over 3 runs with different seeds, error bars indicate standard deviation
:class: sd-align-minor-center | ||
|
||
* The model has ~4.1 million neurons and ~24 billion synapses | ||
* Steady decrease of run time with additional compute resources |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add bullet item:
- data averaged over 3 runs with different seeds, error bars indicate standard deviation
|
||
* The size of network scales proportionally with the computational resources used | ||
* Largest network size in this diagram: ~5.8 million neurons and ~65 billion synapses | ||
* The figure shows that NEST can handle massive networks and simulate them efficiently |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
add bullet item:
- data averaged over 3 runs with different seeds, error bars indicate standard deviation
This PR adds doc page that showcases the performance results of different scaling experiments.
images should be stored in /doc/htmldoc/static/img/
Thanks @ackurth and s. theinen for your work on this
Fixes #3243
Read the Docs
https://nest-simulator--3244.org.readthedocs.build/en/3244/benchmark_results.html