GPUs as an opportunity for offloading garbage collection

Authors:
Martin Maas;Philip Reames;Jeffrey Morlan;Krste Asanović;Anthony D. Joseph;John Kubiatowicz
Affiliations:
University of California, Berkeley, Berkeley, CA, USA;University of California, Berkeley, Berkeley, CA, USA;University of California, Berkeley, Berkeley, CA, USA;University of California, Berkeley, Berkeley, CA, USA;University of California, Berkeley, Berkeley, CA, USA;University of California, Berkeley, Berkeley, CA, USA
Venue:
Proceedings of the 2012 international symposium on Memory Management
Year:
2012

Citing 14
Cited 1

Garbage collection: algorithms for automatic dynamic memory management

Garbage collection: algorithms for automatic dynamic memory management
The Jikes research virtual machine project: building an open-source research community

IBM Systems Journal
The DaCapo benchmarks: java benchmarking development and analysis

Proceedings of the 21st annual ACM SIGPLAN conference on Object-oriented programming systems, languages, and applications
SableVM: a research framework for the efficient execution of java bytecode

JVM'01 Proceedings of the 2001 Symposium on JavaTM Virtual Machine Research and Technology Symposium - Volume 1
Cell GC: using the cell synergistic processor as a garbage collection coprocessor

Proceedings of the fourth ACM SIGPLAN/SIGOPS international conference on Virtual execution environments
Parallel generational-copying garbage collection with a block-structured heap

Proceedings of the 7th international symposium on Memory management
Accelerating large graph algorithms on the GPU using CUDA

HiPC'07 Proceedings of the 14th international conference on High performance computing
Tracing garbage collection on highly parallel platforms

Proceedings of the 2010 international symposium on Memory management
Small-ruleset regular expression matching on GPGPUs: quantitative performance analysis and optimization

Proceedings of the 24th ACM International Conference on Supercomputing
An effective GPU implementation of breadth-first search

Proceedings of the 47th Design Automation Conference
Accelerating CUDA graph algorithms at maximum warp

Proceedings of the 16th ACM symposium on Principles and practice of parallel programming
Iterative data-parallel mark&sweep on a GPU

Proceedings of the international symposium on Memory management
A comprehensive evaluation of object scanning techniques

Proceedings of the international symposium on Memory management
Performance Characterization and Optimization of Atomic Operations on AMD GPUs

CLUSTER '11 Proceedings of the 2011 IEEE International Conference on Cluster Computing

Towards hinted collection: annotations for decreasing garbage collector pause times

Proceedings of the 2013 international symposium on memory management

Quantified Score

Hi-index	0.00

Visualization

Abstract

GPUs have become part of most commodity systems. Nonetheless, they are often underutilized when not executing graphics-intensive or special-purpose numerical computations, which are rare in consumer workloads. Emerging architectures, such as integrated CPU/GPU combinations, may create an opportunity to utilize these otherwise unused cycles for offloading traditional systems tasks. Garbage collection appears to be a particularly promising candidate for offloading, due to the popularity of managed languages on consumer devices. We investigate the challenges for offloading garbage collection to a GPU, by examining the performance trade-offs for the mark phase of a mark & sweep garbage collector. We present a theoretical analysis and an algorithm that demonstrates the feasibility of this approach. We also discuss a number of algorithmic design trade-offs required to leverage the strengths and capabilities of the GPU hardware. Our algorithm has been integrated into the Jikes RVM and we present promising performance results.