Apache Spark is one of the most active open-source cluster computing frameworks. However, there has been very limited research on improving Spark performance using GPUs. This paper explores the feasibility and benefits of off loading Spark core operations to GPUs. Experimental results from popular Spark applications show a performance improvement of up to 17X for a K-Means Clustering application. The results are used to suggest the type of applications that will benefit the most from off loading Spark workloads to GPUs.

In this paper, we investigate an intriguing shifting trend in performance bottlenecks for Near-Threshold Computing (NTC) processors. Our study demonstrates that the traditional memory latency bottleneck is largely superseded by the bottlenecks of Long Latency Datapaths (LLDs) within a processor core. To exploit this paradigm shift, we propose Opportunistic Turbo Execution (OTE). OTE dynamically boosts the performance of LLDs, by several factors, improving both performance and energy efficiency in an NTC core. Using a comprehensive circuit-architectural analysis, we demonstrate a 42.2% improvement in energy efficiency over a recently proposed technique, across a range of benchmarks.