An evaluation of CUDA-enabled virtualization solutions
Authors:
- 1 December 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 621-626
Abstract
Virtualization, as a technology that enables easy and effective resource sharing with a low cost and energy footprint, is becoming increasingly popular not only in enterprises but also in high performance computing. Applications with stringent performance needs often make use of graphics processors for accelerating their computations. Hence virtualization solutions that support GPU acceleration are gaining importance. This paper performs a detailed evaluation of three frameworks: rCUDA, gVirtuS and Xen, which support GPU acceleration through CUDA, within a virtual machine. We describe the architectures of these three solutions and compare and contrast them in terms of their fidelity, performance, multiplexing and interposition characteristics.Keywords
This publication has 6 references indexed in Scilit:
- Enabling CUDA acceleration within virtual machines using rCUDA2011
- Performance of CUDA Virtualized Remote GPUs in High Performance Clusters2011
- vCUDA: GPU-Accelerated High-Performance Computing in Virtual MachinesIEEE Transactions on Computers, 2011
- GPU virtualization on VMware's hosted I/O architectureACM SIGOPS Operating Systems Review, 2009
- GViM2009
- Virtualization for high-performance computingACM SIGOPS Operating Systems Review, 2006