A Type-Oriented Graph500 Benchmark

TL;DR

This paper introduces a type-oriented programming approach for data-intensive HPC workloads, enabling simpler code development and tuning, and evaluates its effectiveness using the Graph500 benchmark compared to MPI implementations.

Contribution

It proposes a novel type-oriented programming model for PGAS-based data-intensive HPC, demonstrating its benefits in programmability and performance tuning.

Findings

Type-oriented implementation simplifies programming effort.

Performance comparable or superior to MPI in Graph500 benchmark.

Type-based tuning enhances data-intensive workload efficiency.

Abstract

Data intensive workloads have become a popular use of HPC in recent years and the question of how data scientists, who might not be HPC experts, can effectively program these machines is important to address. Whilst using models such as Partitioned Global Address Space (PGAS) is attractive from a simplicity point of view, the abstractions that these impose upon the programmer can impact performance. We propose an approach, type-oriented programming, where all aspects of parallelism are encoded via types and the type system which allows for the programmer to write simple PGAS data intensive HPC codes and then, if they so wish, tune the fundamental aspects by modifying type information. This paper considers the suitability of using type-oriented programming, with the PGAS memory model, in data intensive workloads. We compare a type-oriented implementation of the Graph500 benchmark against MPI reference implementations both in terms of programmability and performance, and evaluate how orienting their parallel codes around types can assist in the data intensive HPC field.