Mars

TLDR

MapReduce, originally proposed by Google for large‑scale web search on commodity CPUs, offers a simple programming model, but GPUs—though far more powerful—are difficult to program due to their specialized architecture. This work designs and implements Mars, the first MapReduce framework targeting GPUs, to exploit their computational power for large‑scale data processing. Mars abstracts GPU programming behind the familiar MapReduce API, running on an NVIDIA G80 GPU with over one hundred cores, and is benchmarked against the CPU‑based Phoenix framework. Mars achieves up to a 16× speedup over Phoenix on a quad‑core CPU for six typical web applications.

Abstract

We design and implement Mars, a MapReduce framework, on graphics processors (GPUs). MapReduce is a distributed programming framework originally proposed by Google for the ease of development of web search applications on a large number of commodity CPUs. Compared with CPUs, GPUs have an order of magnitude higher computation power and memory bandwidth, but are harder to program since their architectures are designed as a special-purpose co-processor and their programming interfaces are typically for graphics applications. As the first attempt to harness GPU's power for MapReduce, we developed Mars on an NVIDIA G80 GPU, which contains over one hundred processors, and evaluated it in comparison with Phoenix, the state-of-the-art MapReduce framework on multi-core CPUs. Mars hides the programming complexity of the GPU behind the simple and familiar MapReduce interface. It is up to 16 times faster than its CPU-based counterpart for six common web applications on a quad-core machine.