2011 Poster Sessions : RAMP: A System for Capturing and Tracing Provenance in MapReduce Workflows

Student Name : Hyunjung Park
Advisor : Jennifer Widom
Research Areas: Computer Systems
RAMP (Reduce And Map Provenance) is an extension to Hadoop that supports provenance capture and tracing for MapReduce workflows. RAMP uses a wrapper-based approach, requiring little if any user intervention in most cases, while retaining Hadoop’s parallel execution and fault tolerance. RAMP does not impose heavy time and space overhead during provenance capture and efficiently backward-traces provenance without special indexes.

Hyunjung Park is a Ph.D. candidate in the Department of Electrical Engineering at Stanford University, working with Prof. Jennifer Widom. He received his B.S. from Seoul National University in 2002 and his M.S. from Stanford University in 2009. His research interests include data provenance, cloud computing, and database management systems.