Distributed Computation of Pi with Apache Hadoop

Speaker:	Dr. Tsz-Wo Nicholas SZE
		Yahoo

Title:		"Distributed Computation of Pi with Apache Hadoop"

Date:		Monday, 18 October 2010

Time:		4:00pm - 5:00pm

Venue:		Lecture Theater F (near lifts 25/26), HKUST

Abstract:

In July 2010, we obtained a new record on computing specific bits of the
mathematical constant Pi.  The new record consisted of 256 bits of Pi
around the two quadrillionth bit position.  The computation was carried
out by a MapReduce program called DistBbp.  DistBbp ran on Apache Hadoop
clusters and utilized idle cluster resources without monopolizing the
entire cluster.  One of the largest computations took 23 days of wall
clock time and 503 years of CPU time on a 1000-node cluster.  In this
talk, we discuss our experience in performing such computations and the
lessons we have learned.

Apache Hadoop is an open source distributed computing software system.
It is now the core system behind many companies including Yahoo!,
Facebook, Twitter, etc.. In this talk, we will also give a brief
introduction to Hadoop.

***********************
Biography:

Dr. Tsz-Wo Nicholas Sze received his B.Eng. and M.Phil. degrees in
Computer Science from the Hong Kong University of Science and Technology
in 1999 and 2001, respectively.  He received his Ph.D. degree in Computer
Science from the University of Maryland College Park in 2007.  Afterward,
he joined the Yahoo! Cloud Computing team in Silicon Valley, California,
USA. His research interests include computational number theory and
distributed computing.  He is also a member of the Project Management
Committee of Apache Hadoop, an open source distributed computing software
system.