-
Notifications
You must be signed in to change notification settings - Fork 32
WIP update steps for container tutorial #850
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
base: develop
Are you sure you want to change the base?
Conversation
@stephanielam3211 Once you have the container running, these are the steps we'll want our attendees to work through. |
…into docs-benchpark-container merging tutorial instructions
docs/example-workflow.rst
Outdated
This command does 2 main things. First, it builds all necessary software using Spack. This process may take some time (2-3 minutes). | ||
Second, this command configures files (e.g. Flux submission script) needed to perform the runs that make up the current experiment. | ||
For each run, a directory will be created under ``benchpark/wkp/kripke-benchmark/tutorial/workspace/experiments/kripke/kripke``. | ||
If the setup is successfull, you will see something like this after the setup command:: |
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.
typo here successfull
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.
fixed
docs/example-workflow.rst
Outdated
|
||
benchpark --version | ||
|
||
You should receive ``0.1.0`` on the terminal |
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.
"This should return" instead of "You should receive"
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.
fixed
@michaelmckinsey1 Can you please take another look at the docs page with the additional descriptions? |
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.
Took a full pass
ideally) | ||
**Prerequisites:** | ||
|
||
* Access to a terminal with Benchpark installed (provided automatically by the infrastructure in our `Benchpark Tutorial repository <https://github.com/llnl/benchpark-tutorial>`_) |
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.
Should this be more specific like - You need Docker
, etc?
For Kripke, the default experiment is ``kripke+single_node``. Since single node runs are enabled by default, the ``+single_node`` | ||
variant will be excluded from the commands below. Instead, the instructions below that use experiment specs will | ||
simply use ``kripke`` instead of ``kripke+single_node``. |
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.
+single_node
is mutually exclusive with +strong
, so the below experiment is actually +strong
not single node
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.
For Kripke, we specify the strong scaling experiment on the command line using kripke scaling=strong
.
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.
Then why does Benchpark print kripke+single_node
as an experiment if the actual experiment is kripke scaling=strong
? Something's not adding up between what you're saying and what Benchpark produces, @michaelmckinsey1. I can try rewriting if I understand what the mismatch is.
docs/example-workflow.rst
Outdated
This command does two things. First, it builds all necessary software using Spack. Building the software may take a while to complete, depending | ||
on how many external packages are contained in the system definition from :ref:`Step 3 <step3_label>`. For this tutorial, it should take roughly 5 minutes. |
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.
Not sure if it's necessary to mention externals here, or maybe rephrase. Almost sounds like more externals makes it take longer.
This command does two things. First, it builds all necessary software using Spack. Building the software may take a while to complete, depending | |
on how many external packages are contained in the system definition from :ref:`Step 3 <step3_label>`. For this tutorial, it should take roughly 5 minutes. | |
This command does two things. First, it builds all necessary software using Spack. Building the software may take a while to complete. For this tutorial, it should take roughly 5 minutes. |
2e43f03
to
8049169
Compare
Description
Adding/modifying a system (docs: Adding a System)
systems/system_name/system.py
filesystem_name
in.github/workflows/run.yml
systems/all_hardware_descriptions/hardware_name/hardware_description.yaml
which will appear in the docs catalogueAdding/modifying a benchmark (docs: Adding a Benchmark)
repo/benchmark_name/package.py
plus: create, self-assign, and link here a follow up issue with a link to the PR in the Spack repo.repo/benchmark_name/application.py
plus: create, self-assign, and link here a follow up issue with a link to the PR in the Ramble repo.application.py
or inrepo/benchmark_name/application.py
will appear in the docs catalogueexperiments/benchmark_name/experiment.py
to define a single node and multi-node experiments.github/workflows/run.yml
Adding/modifying core functionality, CI, or documentation:
.github/workflows
and.gitlab/tests
unit tests (if needed)