Dynamic load balancing on single- and multi-GPU systems

The computational power provided by many-core graphics processing units (GPUs) has been exploited in many applications. The programming techniques currently employed on these GPUs are not sufficient to address problems exhibiting irregular, and unbalanced workload. The problem is exacerbated when trying to effectively exploit multiple GPUs concurrently, which are commonly available in many modern systems. In this paper, we propose a task-based dynamic load-balancing solution for singleand multi-GPU systems. The solution allows load balancing at a finer granularity than what is supported in current GPU programming APIs, such as NVIDIA's CUDA. We evaluate our approach using both micro-benchmarks and a molecular dynamics application that exhibits significant load imbalance. Experimental results with a singleGPU configuration show that our fine-grained task solution can utilize the hardware more efficiently than the CUDA scheduler for unbalanced workload. On multiGPU systems, our sol...

• Type: Journal
• Year: 2010
• Where: IPPS
• Authors: Long Chen, Oreste Villa, Sriram Krishnamoorthy, Guang R. Gao
• Keywords: Distributed And Parallel Computing, GPUs, IPPS 2010, Solution, Unbalanced Workload