Componentizing Ansys packages#

Componentization is the process of subdividing the functionality of large apps into multiple self-contained services with independent APIs. API creation surrounding existing Ansys products naturally aligns to publishing packages that mimic the full domain and scope of each product.

Emphasizing component libraries and services during API exposure sets a new paradigm for Ansys product architecture that inherently breaks apart larger monolithic desktop apps into subsets of functionality, with the expectation of compatibility and reusability across the entire Ansys portfolio.

Many Ansys products already have a scripting solution in place, and wrapping that execution environment with a RunScript API endpoint is a low-barrier option to gain access to remote, programmatic execution. This solution lacks API granularity, however, as the abstraction is simply an unvalidated script input and some blob output that must be parsed and evaluated without a prescribed response definition.

The documentation for the scripting environment still remains deep within the product and each script execution request is hard to organize and maintain. Thus, there remains a significant cognitive disconnect when consuming this API abstraction. The lack of API definition within the top-level abstraction also makes data management difficult and direct compatibility with other PyAnsys libraries challenging.

In addition to API clarity, the underlying product keeps a very large installation footprint that is a burden in modern, flexible cloud deployments. While avoiding re-architecting a product in the short-term can give a quick win, in most cases, it should only be used as a stopgap solution, providing a window of opportunity to learn more about how the user prefers to consume the individual functionalities of a product.

Creating well-architected component libraries maximizes product usage with these key benefits:

  • Reusable shared components

  • Improved API quality

  • Package size reduction

  • Product compatibility and composability

  • Optimized, on-demand user solutions