OptimusPy for TM1

Find the ideal dimension order for your TM1 cubes

OptimusPy finds a better storage dimension order for a TM1 cube by applying real reorders on the server and measuring memory, query time and TI time for each. It reports the best order and can apply it for you. It supports TM1 v11 and v12 (PAoC/PAaaS).

Setup

1. Install. Either:

   • download the Windows (optimuspy-windows.zip) or Linux (optimuspy-linux.tar.gz, or optimuspy-linux-arm64.tar.gz on ARM) bundle from the Releases page and unpack it. It holds the executable, config/config.ini.example and samples/ (example cube configs). No Python needed;
   • or, with Python 3.11 or later, clone the repository and run pip install -e ., which adds the optimuspy command.

   See Installation.

2. Connect to TM1. Copy config/config.ini.example to config/config.ini and add a section for your instance. If you already keep a config.ini for RushTI or your scripts, link it on the UI's Settings page instead. See TM1 Connection.

Run it in the UI

optimuspy ui

Or double-click the executable. The UI opens at http://127.0.0.1:8765. Then:

1. Choose your instance in the sidebar.
2. On the Optimize page, click a cube.
3. On Configure, keep Greedy, pick a view and click Save & Start Optimization.
4. When it ends, open the HTML report from the cube's Reports tab.

Run it on the command line

Describe the cube in a JSON file, like samples/optimize.json:

{
  "instance": "tm1srv01",
  "cube": "Sales",
  "views": ["Optimus"],
  "processes": [],
  "executions": 10,
  "fast": false,
  "output": "csv",
  "update": false,
  "dimensions_to_exclude": [],
  "predefined_orders": [],
  "orders_to_ignore": []
}

Then run it:

optimuspy optimize samples/optimize.json

The report is written to results/<instance>/. With "update": false the cube ends in its original order.

To find which cubes are worth optimizing, scan the instance. This lists the largest cubes, the ones that make up 60% of its memory, leaves out those already optimized, and writes a starter config for each into cube-configs/; add your views to each file before you run it:

optimuspy scan --instance tm1srv01 --output cube-configs/

Documentation

cubewise-code.github.io/optimus-py/docs

• New to OptimusPy? Start with the User Guide.
• What's new in 2.0.0, and how to upgrade from 1.x: CHANGELOG.md.
• Contributing, and how versions and releases are made: CONTRIBUTING.md.

License

This project is licensed under the MIT License - see the LICENSE file for details