In this paper, Takizawa et al. have presented a tool named CheCUDA that is designed to checkpoint CUDA applications. As existing checkpoint/restart implementations do not support checkpointing the GPU status, CheCUDA hooks basic CUDA driver API calls in order to record the GPU status changes on the main memory. At checkpointing, CheCUDA stores the status changes in a file after copying all necessary data in the video memory to the main memory and then disabling the CUDA runtime. At restart, CheCUDA reads the file, re-initializes the CUDA runtime, and recovers the resources on GPUs so as to restart from the stored status. This paper demonstrates that a prototype implementation of CheCUDA can correctly checkpoint and restart a CUDA application written with basic APIs. This also indicates that CheCUDA can migrate a process from one PC to another even if the process uses a GPU. Accordingly, CheCUDA is useful not only to enhance the dependability of CUDA applications but also to enable dynamic task scheduling of CUDA applications required especially on heterogeneous GPU cluster systems. This paper also shows the timing overhead for checkpointing.

(Hiroyuki Takizawa, Katuto Sato, Kazuhiko Komatsu, and Hiroaki Kobayashi, CheCUDA: A Checkpoint/Restart Tool for CUDA Applications, to appear inProceedings of the Tenth International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT) 2009, Workshop on Ultra Performance and Dependable Acceleration Systems).