Extending and Implementing the Self-adaptive Virtual Processor for Distributed Memory Architectures

TL;DR

This paper extends the Self-adaptive Virtual Processor (SVP) model to support distributed memory architectures, enabling fine-grained concurrency management across heterogeneous clusters while maintaining the original programming interface.

Contribution

The paper introduces extensions to SVP for distributed environments and presents a prototype implementation supporting heterogeneous clusters over TCP/IP.

Findings

Prototype supports execution on distributed memory clusters

Maintains original SVP programming model in distributed settings

Enables fine-grained concurrency management across distributed systems

Abstract

Many-core architectures of the future are likely to have distributed memory organizations and need fine grained concurrency management to be used effectively. The Self-adaptive Virtual Processor (SVP) is an abstract concurrent programming model which can provide this, but the model and its current implementations assume a single address space shared memory. We investigate and extend SVP to handle distributed environments, and discuss a prototype SVP implementation which transparently supports execution on heterogeneous distributed memory clusters over TCP/IP connections, while retaining the original SVP programming model.