Extending Abstract GPU APIs to Shared Memory
Date:
Parallel programming is widely used for general-purpose computations, but the performance of different parallel APIs often varies depending on the problem type and hardware architecture. This variation creates a need for an abstract representation to express parallel problems independently of specific platforms.
This work presents a new approach that allows programmers to utilize various parallel APIs without focusing on their technical or platform-specific details. Building on our earlier Abstract Application Programming Interface (API) designed for Graphical Processing Unit (GPU) programming, we extend the concept to support shared memory systems using OpenMP.