Open Analytics Framework is a scalable and extensible framework for executing third-party open source and custom analytics.
A brief schematic is illustrated in the following diagram.
- Analytic workloads separation by running scripts in containers
In this way, containers can be also used or leveraged by different Vantage features.
- Third-party software and open source languages, such as Python
- Secure, custom user environments
- Elastic scaling
- Superior price performance
The Open Analytics Framework can be set up and operated with the analytic compute groups of VantageCloud Lake. See How Open Analytics Framework Works section for an overview of how the framework operates.
The key database component that enables running user scripts in the Open Analytics Framework is the new APPLY table operator. This operator is only available in the Database Engine 20, and is replacing the SCRIPT and ExecR table operators. The latter objects will continue to operate in VantageCloud Enterprise and VantageCore systems.
- Develop analytics programs with first class support in Python.
- Install code and their dependent libraries or binaries.
- For Python, from the Python Package Index (PyPI) repository of the Python Software Foundation (PSF).
- Use the Database Engine 20 APPLY table operator to execute analytics in parallel on analytic compute groups.
- Process Database Engine 20 database rows as input to their analytic workloads.
- Use the teradataml Python client analytic library APIs as primary interface to perform all tasks.
The Open Analytics Framework is a pre-configured component for all VantageCloud Lake systems. It runs exclusively on analytic compute group clusters and can autoscale as part of the VantageCloud Lake architecture.