Virtual Machine Support for Many-Core Architectures: Decoupling Abstract from Concrete Concurrency Models

Abstract

The upcoming many-core architectures require software developers to exploit concurrency to utilize available computational power. Today's high-level language virtual machines (VMs), which are a cornerstone of software development, do not provide sufficient abstraction for concurrency concepts. We analyze concrete and abstract concurrency models and identify the challenges they impose for VMs. To provide sufficient concurrency support in VMs, we propose to integrate concurrency operations into VM instruction sets. Since there will always be VMs optimized for special purposes, our goal is to develop a methodology to design instruction sets with concurrency support. Therefore, we also propose a list of trade-offs that have to be investigated to advise the design of such instruction sets. As a first experiment, we implemented one instruction set extension for shared memory and one for non-shared memory concurrency. From our experimental results, we derived a list of requirements for a full-grown experimental environment for further research

Many-core virtual machines

“…Our main goal is to identify common abstractions of different abstract concurrency models which are also appropriate to be mapped efficiently onto the various upcoming many-core architectures [4]. Thus, the concurrency models on the different levels of implementation are to be decoupled.…”

“…Thus, we investigated the state of the art in concurrency support for virtual machines [3] and experimented with support for threads and locks, as well as actor abstractions on the instruction set level [4]. Our experiments included also an analysis of high-level concurrency constructs with a focus on barrier synchronization.…”

Challenging VMs on battery-powered embedded devices

“…The Renaissance project by David Ungar and Sam Adams proposes a virtual machine able to use 56 of the 64 Tilera cores 1 . We also found work of Stefan Marr on adding concurrency support in instruction sets of virtual machines (Marr et al 2009). There is also rumors of an Erlang port on the Tilera64 2 .…”

RELEASE: A High-Level Paradigm for Reliable Large-Scale Server Software

“…Such models must abstract from variations introduced by differences between multi-core architectures and uniformly treat different hardware architectures, number of cores and memory access characteristics. Currently, the implementation of such models in the form of high-level languages and libraries is not sufficient; these two must be complemented with efficient and reliable virtual machines (VMs) that provide inherent support for high-level concurrency [13].…”