Extended abstract: Type oriented programming for task based parallelism

TL;DR

This paper proposes a type-oriented programming approach that allows programmers to annotate code with type information, enabling a balance between high-level abstraction and low-level performance optimization in task-based parallelism.

Contribution

It introduces a novel method where optional type annotations guide compilers to optimize parallel code, bridging the gap between ease of programming and performance.

Findings

Type annotations improve parallel code performance.

The approach simplifies parallel programming without sacrificing scalability.

Compiler-guided optimizations lead to better resource utilization.

Abstract

Writing parallel codes is difficult and exhibits a fundamental trade-off between abstraction and performance. The high level language abstractions designed to simplify the complexities of parallelism make certain assumptions that impacts performance and scalability. On the other hand lower level languages, providing many opportunities for optimisation, require in-depth knowledge and the programmer to consider tricky details of parallelism. An approach is required which can bridge the gap and provide both the ease of programming and opportunities for control and optimisation. By optionally decorating their codes with additional type information, programmers can either direct the compiler to make certain decisions or rely on sensible default choices.