Examples¶
Examples |
Parfun |
Pargraph |
Client |
Workers |
Num Workers |
Ratio: Speed/Workers |
Sequential Runtime |
Parallel Runtime |
Speedup |
|---|---|---|---|---|---|---|---|---|---|
Yes |
No |
AWS |
EC2 |
8 |
0.31 |
14m 38s |
5m 49s |
2.51 |
|
Yes |
No |
AWS |
EC2 |
128 |
0.26 |
81m 46s |
2m 12s |
33 |
|
Yes |
Yes |
AWS |
EC2 |
64 |
0.43 |
35m 12s |
1m 16s |
27.6 |
|
No |
Yes |
NATIVE |
NATIVE |
16 |
0.64 |
64m 04s |
6m 14s |
10.27 |
Running these notebooks locally¶
The notebooks live under examples/notebooks/ in the
OpenGRIS Scaler repository. From a
fresh checkout, set them up in four steps:
Set up your environment. Create and activate a virtual environment, then install OpenGRIS Scaler (see Installation for more options):
uv venv source .venv/bin/activate uv pip install opengris-scaler
Install the notebook dependencies from the bundled requirements file:
uv pip install -r examples/notebooks/requirements_notebooks.txtInstall Jupyter:
uv pip install jupyterLaunch Jupyter and run the code:
cd examples/notebooks jupyter notebook
Open a notebook and run its cells in order. Notebooks that connect to a Scaler cluster over a
tcp://address expect a running scheduler and cluster; the others start a local cluster automatically.