Mars: Accelerating MapReduce with Graphics Processors

MPhil Thesis Defence


Title: "Mars: Accelerating MapReduce with Graphics Processors"

By

Mr. Wenbin Fang


Abstract

We design and implement Mars, a MapReduce runtime system accelerated with 
graphics processing units (GPUs). MapReduce is a simple and flexible parallel 
programming paradigm originally proposed by Google, for the ease of large scale 
data processing on thousands of CPUs. Compared with CPUs, GPUs have an order of 
magnitude higher computation power and memory bandwidth. However, GPUs are 
designed as special-purpose co-processors and their programming interfaces are 
less familiar than those on the CPUs to MapReduce programmers.

To harness GPUs power for MapReduce, we developed Mars to run on NVIDIA GPUs, 
AMD GPUs, as well as multi-core CPUs. Furthermore, we integrated Mars into 
Hadoop, an open-source CPU-based distributed MapReduce system. Mars hides the 
programming complexity of GPUs behind the simple and familiar MapReduce 
interface, and automatically manages task partitioning, data distribution, and 
parallelization on the processors. We have implemented six representative 
applications on Mars and evaluated their performance on PCs equipped with GPUs 
as well as multi-core CPUs. The GPU acceleration with an NVIDIA GTX280 achieved 
a speedup of an order of magnitude over a quad-core CPU. Utilizing both the GPU 
and the CPU further improved GPU-only performance by 40% for some applications. 
Additionally, integrating Mars into  Hadoop enabled GPU acceleration for a 
network of PCs.


Date:			Friday, 25 June 2010

Time:			10:00am - 12:00noon

Venue:			Room 3501
 			Lifts 25/26

Committee Members:	Dr. Qiong Luo (Supervisor)
 			Dr. Shing-Chi Cheung (Chairperson)
 			Dr. Lin Gu


**** ALL are Welcome ****