Memory-Side Acceleration for XML Parsing
Source: University of California
As Extensible Markup Language (XML) becomes prevalent in cloud computing environments, it also introduces significant performance overheads. In this paper, the authors analyze the performance of XML parsing, identify that a significant fraction of the performance overhead is indeed incurred by memory data loading. To address this problem, they propose implementing memory-side acceleration on top of computation-side acceleration of XML parsing. To this end, they study the impact of memory-side acceleration on performance, and evaluate its implementation feasibility including bus bandwidth utilization, hardware cost, and energy consumption. Their results show that this technique is able to improve performance by up to 20% as well as produce up to 12.77% of energy saving when implemented in 32 nm technology.