The goal of my research
is to develop practical techniques for distributed systems
to improve scalability, security, and reliability.
I'm particularly interested in
data management and networked systems:
large-scale distributed systems, database systems, operating systems, and
security. My thesis research is on designing and building
scalable and robust services for distributed monitoring.
In prior work,
I have built
an adaptive resource manager and performance
benchmarks for distributed stream processing,
a resource allocation framework to achieve high availability for
distributed applications,
and a
scalable replica synchronization protocol for data replication.
I'm currently working on the following projects:
-
My Erdös number (shortest academic path to Paul Erdös) is
four
-
System and Method for Resource Allocation in Partial Fault-Tolerant Applications.
Navendu Jain, Yoonho Park, Deepak Turaga, and Chitra Venkatramani.
YOR920070525US1, Filed 2008.
-
A Method for Redundancy Elimination for Replicating File System Data.
Navendu Jain and Renu Tewari.
ARC-820040139, 2004.
-
Two New Content-based Techniques for Similarity Detection.
Navendu Jain and Renu Tewari.
ARC-820040156, 2004.
-
SMART: Scalable, Bandwidth-Aware Monitoring of Continuous Aggregation Queries.
Navendu Jain, Praveen Yalagandula, Mike Dahlin, and Yin Zhang.
[PDF]
[PS]
[Bibtex]
[Project Page]
-
WebView: Scalable Information Monitoring for Data-Intensive Web Applications.
Navendu Jain, Mike Dahlin, and Yin Zhang.
[PDF] [PS]
[Bibtex]
[Project Page]
-
Network Imprecision: A New Consistency Metric for Scalable Monitoring.
Navendu Jain, Dmitry Kit, Prince Mahajan, Praveen Yalagandula, Mike Dahlin, and Yin Zhang.
[PDF] [PS]
[Technical Report]
[Bibtex]
[Project Page]
-
Towards Optimal Resource Allocation in Partial-Fault Tolerant Applications.
Nikhil Bansal, Ranjita Bhagwan, Navendu Jain, Yoonho Park, Deepak Turaga, and Chitra Venkatramani.
27th IEEE International Conference on Computer Communications (INFOCOM '08)
April 2008 (To Appear). Acceptance Ratio: 20.5% (236 out of 1152)
[PDF]
[PS]
[Bibtex]
[Technical Report]
[Project Page]
-
STAR: Self-Tuning Aggregation for Scalable Monitoring.
Navendu Jain, Dmitry Kit, Prince Mahajan, Praveen Yalagandula, Mike Dahlin, and Yin Zhang.
33rd International Conference on Very Large Databases (VLDB '07)
Vienna, Austria, September 2007. Acceptance Ratio: 16.4% (45 out of 275)
[PDF] [PS]
[Bibtex]
[Technical Report]
[Project Page]
-
Design, Implementation, and Evaluation of the Linear Road Benchmark on the Stream Processing Core.
Navendu Jain, Lisa Amini, Henrique Andrade, Richard King, Yoonho Park, Philipe Selo, Chitra Venkatramani.
25th ACM SIGMOD International Conference on Management of Data (SIGMOD '06)
Chicago, IL, June 2006. Acceptance Ratio: 13% (58 out of 446)
[PDF]
[PS]
[Bibtex]
[Project Page]
-
Adaptive Control of Extreme-Scale Stream Processing Systems.
Lisa Amini, Navendu Jain, Anshul Sehgal, Jeremy Silber, Olivier Verscheure.
26th International Conference on
Distributed Computing Systems (ICDCS '06)
Lisboa, Portugal, July 2006. Acceptance Ratio: 14% (75 out of 540)
[PDF]
[PS]
[Bibtex]
[Project Page]
-
TAPER: Tiered Approach for Eliminating Redundancy in Replica Sychronization.
Navendu Jain, Mike Dahlin, and Renu Tewari.
4th USENIX Conference on File and Storage Technologies (FAST '05)
San Francisco, CA, December 2005. Acceptance Ratio: 20% (25 out of 125)
[PDF]
[PS]
[Bibtex]
[Project Page]
-
INSIGHT: A Distributed Monitoring System for Tracking Continuous Queries.
Navendu Jain, Praveen Yalagandula, Mike Dahlin, and Yin Zhang.
20th ACM Symposium on Operating Systems Principles, Work-in-Progress (SOSP '05 WIP)
Brighton, United Kingdom, October 2005.
[PDF]
[PS]
[Bibtex]
[Slides]
[Project Page]
-
Scaling Real-Time Telematics Applications using Programmable Middleboxes.
Annie Chen, Navendu Jain, Tadeusz Pietraszek, Angelo Perniola, Sean
Rooney, and Paolo Scotton.
IEEE
Consumer Communications and Networking 2004 (CCNC '04)
Las Vegas, Nevada, January 2004.
[PDF]
[PS]
[Bibtex]
-
An Architectural Framework to deploy
Scatternet-based Applications over Bluetooth.
Nitin Pabuwal, Navendu Jain, and B. N. Jain.
IEEE
International Conference on Communications (ICC '03)
Anchorage, Alaska, May 2003.
[PDF]
[PS]
[Bibtex]
-
Verification of Timed Automata via Satisfiability Checking.
P. Niebert, M. Mahfoudh, Eugene Asarin, Marius Bozga, Navendu Jain and Oded Maler.
7th International Symposium
on Formal Techniques in Real-Time and Fault Tolerant Systems
(FTRTFT'02)
Oldenburg, Germany, September 2002.
[PDF]
[PS]
[Bibtex]
-
Improving Image Retrieval Performance using Negative Relevance Feedback.
T.V. Ashwin, Navendu Jain, and Sugata Ghosal
IEEE International Conference on Acoustics, Speech and Signal
Processing (ICASSP '01)
Salt Lake City, Utah, May 2001.
[PDF]
[PS]
[Bibtex]
-
Efficient Object Ellipsoid Trees.
Navendu Jain, Sorav Bansal, and Sanjiv Kapoor
Indian
Conference on Computer Graphics, Vision and Image Processing (ICVGIP '00)
Bangalore, India, December 2000.
[PDF]
[PS]
[Bibtex]
Refereed Workshop Papers
-
Scatternet Formation for Multimedia Applications
over Bluetooth Personal Area Networks.
Navendu Jain.
M.Tech. thesis, IIT Delhi.
May 2002.
[PDF] [PS]
[Bibtex]
-
PRISM: PRecision-Integrated Scalable Monitoring.
Navendu Jain, Dmitry Kit, Prince Mahajan, Praveen Yalagandula, Mike Dahlin, and Yin Zhang.
TR06-22, UTCS Technical Report.
[PDF]
[PS]
[Bibtex]
[Project Page]
[PRISMon Demo]
-
Distributed, Robust, and Self-Organizing Bluetooth Scatternet Formation.
Navendu Jain, Nitin Pabuwal, and B. N. Jain.
TR04-41, UTCS Technical Report.
January 2004.
[PDF] [PS]
-
The GPU as a Scientific Computing Engine: Performance and Analysis.
Navendu Jain and Jason Chaw.
TR04-40, UTCS Technical Report.
January 2004.
[PDF] [PS]
-
Robust Non-parametric Relevance Feedback for Image Retrieval.
Ashwin T.V, Sugata Ghosal , Abhinanda Sarkar, Navendu Jain and Susmit Sarkar
IBM IRL Technical Report.
December 2001.
[PDF]
[PS]
-
PRISM/SDIMS Monitoring System: PRISM
is a research project on designing and building a scalable, high performance,
and robust monitoring framework for large-scale networked systems. PRISM
is implemented on top of the SDIMS aggregation system using FreePastry.
It is being currently used by researchers at
UCSD, Univ. of Arizona, and UIUC. We are planning a public release but
please contact me if you are interested.
-
Linear Road Benchmark:
Linear Road is a benchmark to
compare performance characteristics of
stream processing engines (SPEs) relative to each other
and relative to classic data management systems.
I designed and implemented this benchmark on IBM's SPC system
The source code is now a part of IBM Research System S project.
Please contact me if you are interested.
-
TAPER data replication:
TAPER is a scalable data replication protocol that
synchronizes a large collection of data across multiple
geographically distributed replica locations.
A prototype implementation of TAPER is available here.