Proof-of-Concept Examples of Performance-Transparent Programming Models

TL;DR

This paper presents proof-of-concept programming models that make machine-specific performance details transparent, enabling better understanding and optimization of applications' memory, vector utilization, and data reuse.

Contribution

It introduces performance-transparent programming models with concrete examples that reveal hardware performance metrics, a novel approach to exposing machine details in programming.

Findings

Memory footprint prediction accuracy up to 25%

Revealed vector unit utilization and data reuse patterns

Demonstrated feasibility of performance transparency in programming models

Abstract

Machine-specific optimizations command the machine to behave in a specific way. As current programming models largely leave machine details unexposed, they cannot accommodate direct encoding of such commands. In previous work we have proposed the design of performance-transparent programming models to facilitate this use-case; this report contains proof-of-concept examples of such programming models. We demonstrate how programming model abstractions may reveal the memory footprint, vector unit utilization and data reuse of an application, with prediction accuracy ranging from 0 to 25 \%.