Publications
A Case for Machine Learning to Optimize Multicore Performance.
Workshop on Hot Topics in Parallel Computing (HotPar), March 2009.
Workshop on Hot Topics in Parallel Computing (HotPar), March 2009.
Predicting Multiple Performance Metrics for Queries: Better Decisions Enabled by Machine Learning. International Conference on Data Engineering (ICDE) March 2009 (to appear).
Managing Operational Business Intelligence Workloads.
SIGOPS Operating Systems Review, Volume 43, No 1, January 2009, pp 92-98.
SIGOPS Operating Systems Review, Volume 43, No 1, January 2009, pp 92-98.
Predicting Query and Workload Performance for Very Large Data Warehouses.
Enterprise Computing and Systems Research (ECSR) Workshop, December 2007.
Enterprise Computing and Systems Research (ECSR) Workshop, December 2007.
Tools and Techniques for Failure Data Collection and Analysis.
Workshop on Reliability Analysis of Systems Failure Data (RAF-2007).
Workshop on Reliability Analysis of Systems Failure Data (RAF-2007).
Windows XP Kernel Crash Analysis.
20th Large Installation System Administration Conference (LISA-2006).
20th Large Installation System Administration Conference (LISA-2006).
Crash Data Collection: A Windows Case Study.
International Conference on Dependable Systems and Networks (DSN-2005).
International Conference on Dependable Systems and Networks (DSN-2005).
Why PCs are Fragile and What We Can Do About It: A Study of Windows Registry Problems. International Conference on Dependable Systems and Networks (DSN-2004).
Why do Internet services fail, and what can be done about it?
4th USENIX Symposium on Internet Technologies and Systems (USITS 2003).
4th USENIX Symposium on Internet Technologies and Systems (USITS 2003).
Refereed Conference Papers
Unrefereed Technical Reports
Why Does Windows Crash?
Computer Science Master’s Thesis, University of California, Berkeley. UCB//CSD-05-1393
Why PCs are Fragile and What We Can Do About It: A Study of Windows Registry Problems (Extended Version), Microsoft Corporation, MSR-TR-2004-25
Failure Analysis of Internet Services. Computer Science Undergraduate Honors Thesis, University of California, Berkeley. UCB//CSD-03-1255
Using Machine Learning to Auto-tune a Stencil Code on a Multicore Architecture.
3rd Workshop on Tackling Systems Problems with Machine Learning Techniques (SysML), Dec 2008.
3rd Workshop on Tackling Systems Problems with Machine Learning Techniques (SysML), Dec 2008.
Characterizing Proprietary Workload to Evaluate, Replay and Predict System Behavior.
21st ACM Symposium on Operating Systems Principles (SOSP), October 2007.
Characterizing and Replaying Proprietary Workloads for Evaluating Systems.
4th USENIX Symposium on Networked Systems Design and Implementation (NSDI), April 2007.
Supporting Annotation Layers for Natural Language Processing.
BioLink 2004: Linking Biological Literature, Ontologies and Databases: Tools for Users, May 2004.
Conference Posters
Predicting Multiple Performance Metrics for Queries: Better Decisions Enabled by Machine Learning. Submitted to 34th International Conference on Very Large Data Bases (VLDB), 2008.
Predicting Query and Workload Performance for Very Large Data Warehouses. Submitted to SIGMOD: Special Interest Group on Management of Data, 2008.
A Case for DataForge: A SourceForge For Experimental Data. Submitted to HotOS X: Tenth Workshop on Hot Topics in Operating Systems, 2005.
Supporting Annotation Layers for Natural Language Processing. Submitted to 42nd Annual Meeting of the Association for Computational Linguistics, 2004.
Socio-Cultural Dynamics of Indian Classical Dance. South Asian Studies Honors Thesis, University of California, Berkeley.
Other Reports
