Behavioural types for non-uniform memory accesses

TL;DR

This paper introduces a type and effect system for actor-based programs on NUMA architectures, enabling analysis of data placement and remote memory access patterns to improve parallel performance.

Contribution

It proposes a novel type system combining ownership and behavioural types to model topology and remote memory effects in NUMA systems.

Findings

Type system accurately models data placement and remote reads/writes.

Framework helps optimize concurrent execution on NUMA architectures.

Enhances understanding of performance impacts of memory access patterns.

Abstract

Concurrent programs executing on NUMA architectures consist of concurrent entities (e.g. threads, actors) and data placed on different nodes. Execution of these concurrent entities often reads or updates states from remote nodes. The performance of such systems depends on the extent to which the concurrent entities can be executing in parallel, and on the amount of the remote reads and writes. We consider an actor-based object oriented language, and propose a type system which expresses the topology of the program (the placement of the actors and data on the nodes), and an effect system which characterises remote reads and writes (in terms of which node reads/writes from which other nodes). We use a variant of ownership types for the topology, and a combination of behavioural and ownership types for the effect system.