|
Sanjay Agrawal Resume: [Word format]
|
·
Conference Management
Toolkit 2.0(CMT) (Aug 2006 – present)
http://cmt.research.microsoft.com/cmt/
The
Conference Management Service (CMT) 2.0 is a conference management service
sponsored by Microsoft Research, and uses the latest Microsoft
technologies. It is capable of handling the complex workflow of an
academic conference including paper submission, reviewing, bidding and online
publishing. Since 2006, I have been
primarily responsible for design and managing development of CMT. The project
involves planning the software development life cycle, setting milestones for
the deliverables and leading the team to achieve the deliverables in a timely
and responsible manner.
·
Richer
Search techniques on MSN live (Aug 2006 – present)
http://research.microsoft.com/dmx/Data_Exploration
The
emphasis is to identify techniques to enhance the user search experience. As
part of the project I have worked on techniques to perform efficient entity
extraction on documents, and how it can be leveraged to provide users with much
richer search experience.
·
Autoadmin
(Feb 1999 –Aug 2006)
http://www.research.microsoft.com/dmx/autoadmin
The
goal of the project is to make databases self tuning and administering. As part
of this project I have worked on Database physical design tuning technologies
in Microsoft SQL Server (SQL Server 2005 and SQL Server 2000). I worked with
other group members towards the research, design, and planning of this product
across different releases of SQL Server.
v Database Tuning Advisor
(DTA)
DTA is currently shipped as a
part of the SQL server 2005 and 2008 release. As part of 2005 release of this
tool, in addition to the core development of the tool, I worked on algorithms
including integration of horizontal partitioning as part of physical design
selection. I developed techniques to enable scaling of the tool to large
workloads and complex databases.
v Index Tuning Wizard (ITW)
ITW is part
of SQL server 7.0 and 2000 release. ITW was the first tool in the industry to
provide automatic index and indexed view recommendations for relational
databases while leveraging history of database usage. I worked on design and
development of core techniques that we used to integrate index and indexed view
selection.
·
DBXplorer:
Extending keyword search to databases (Apr 2000 – Dec 2003)
http://research.microsoft.com/dmx/Data_Exploration
The
objective of the project is to extend the universal “keyword search” paradigm
to databases. The project mixes the powerful keyword search with navigational
capabilities to provide users with a seamless way to navigate data that resides
in relational databases, without having to know the underlying schema.
Challenges include approximate answers and ranking of result sets to provide an
IR like interface to databases.
20 Patents in the area of physical database design,
ranking of query results and search over databases.
Contact Info:
Microsoft Research,
One Microsoft Way, Microsoft Corp.,
Redmond, WA 98052 USA
Telephone (425) 705 3507 Fax (425) 936 7329 Email
sagrawal@microsoft.com