MIT: Independent Activities Period: IAP

IAP 2013



Tradeoffs between Massively Parallel Analytics Systems

Andrew Lamb, MIT EECS SB/MENg

Jan/10 Thu 06:00PM-07:00PM 32-141

Enrollment: Unlimited: No advance sign-up
Prereq: none

In this talk, we will enumerate some of the many tradeoffs
between the various analytical systems available for processing extremly
large data sets.  We will compare Vertica and other parallel database
systems, Hadoop/MapReduce, HBase/Cassandra, and Pig, Hive/Impala.

Speaker:
Andrew Lamb (aalamb@alum.mit.edu), MIT Course VI 2002, MEng 2003. After
graduating from MIT with an MEng thesis focused on compilers, he written
system software for Oracle, DataPower/IBM and Vertica/HP for over 9
years. During his last few years at Vertica he has seen and experienced
first hand the joy and challenge of processing massive amounts of data
with grids of commodity servers and has definite opinions on the
strengths and weaknesses of various approaches.

 

 

Sponsor(s): Electrical Engineering and Computer Science
Contact: Andrew Lamb, alamb@vertica.com