Laura Haas
IBM Distinguished Engineer
Director of Computer Science at Almaden Research Center

Wednesday, March 28, 2007
EMCB 101
Time: 3:20 pm
Host: Juliana Freire

Beauty and the Beast: The Theory and Practice of Information Integration

Abstract
Information integration is becoming a critical problem for businesses and individuals alike. Data volumes are sky-rocketing, and new sources and types of information are proliferating. This paper briefly reviews some of the key research accomplishments in information integration (theory and systems), then describes the current state-of-the-art in commercial practice, and the challenges (still) faced by CIOs and application developers. One critical challenge is choosing the right combination of tools and technologies to do the integration. Although each has been studied separately, we lack a unified (and certainly, a unifying) understanding of these various approaches to integration. Experience with a variety of integration projects suggests that we need a broader framework, perhaps even a theory, which explicitly takes into account requirements on the result of the integration, and considers the entire end-to-end integration process.

Bio
Laura Haas is an IBM Distinguished Engineer and Director of Computer Science at Almaden Research Center. Most recently, she led the Information Integration Solutions technical team in IBM's Software Group. Previously, Dr. Haas was a research staff member and manager at Almaden. She is best known for her work on the Starburst query processor (from which DB2 UDB was developed), on Garlic, a system which allowed federation of heterogeneous data sources, and on Clio, the first semi-automatic tool for heterogeneous schema mapping. Dr. Haas is Vice President of the VLDB Board of Trustees, a member of the IBM Academy of Technology, and an ACM Fellow.

---

Kathryn McKinley
Professor
Department of Computer Sciences, University of Texas at Austin

Wednesday, April 18, 2007
EMCB 101
Time: 3:20 pm
Host: Women@CS

Schedule

Dynamic Languages: Opportunities and Challenges

Abstract
Programmers are increasingly turning to managed languages, such as Java and C#, because the resulting programs are faster to implement and have fewer defects. However, many architects and systems researchers do not yet evaluate or design new approaches for these workloads, retarding innovation. One reason may be that measuring dynamic languages has a number of challenges, such as how to factor in and out dynamic compilation and garbage collection. In the first half of the talk, we show (1) the pitfalls of ignoring these effects, and (2) how to measure managed languages in a meaningful way. In the second half of the talk, we show that managed languages offer the potential for coupling high productivity with high performance compared with conventional languages such as C. We overview a few novel cooperative compiler and runtime optimizations, such as dynamic object layout, object colocation, and compiler inserted frees for incremental reclamation.

Bio
Professor Kathryn S. McKinley is a David Bruton Jr. Centennial Fellow at the University of Texas at Austin. She received her PhD from Rice University in 1992. Her research interests include compilers, architecture, and distributed information retrieval. Her most recent projects include automatic and explicit memory management systems, and compiling for TRIPS, a next generation scalable architecture. Her honors include three IBM Faculty Awards (2003, 2004, 2005), College of Natural Science's Dean's Fellow (2005-2006), an NSF CAREER Award (1996-2000), IEEE Senior member, a Texas Institute for Computation and Applied Mathematics (TICAM) Visitor Fellowship (1999-2000), a Chateaubriand Scholarship for the Exact Sciences, Engineering, and Medicine (1992-1993), and a DARPA/NASA Assistantship in Parallel Processing (1991).

---

Susan Davidson
Professor
Department of Computer and Information Sciences, University of Pennsylvania

Monday, September 17, 2007
MEB 3147
Refreshments: 3:20 pm
Lecture 3:40 pm

CRA-W Panel: Graduate School
11:15 am
MEB 3105
Host: Juliana Freire

Sponsored by CRA-W

Querying and Managing Provenance though User Views in Scientific Workflows

Abstract
Workflow systems have become increasingly popular for managing large-scale in-silico experiments where many bioinformatics tasks are chained together. Due to the large amount of data products generated by these experiments and the need for reproducible results, provenance has become of paramount importance. Several workflow systems are therefore starting to provide support for querying provenance. However, the amount of provenance information produced may be overwhelming, so there is a need for abstraction mechanisms to present the most relevant information.

The technique we pursue is that of "user views." Since bioinformatics tasks may themselves be complex sub-workflows, the notion of a user view determines what level of granularity the user can see in the workflow. For example, biologists may simply wish a view in which reformatting tasks are hidden and biologically relevant tasks are seen. Thus the user view determines what data products and tasks can be seen and queried when answering questions of provenance. This talk gives an example of a phylogenomic analysis workflow, discusses the notion of user views relative to this workflow, demonstrates how user views can be used in provenance queries, and discusses how a user view is generated based on what tasks the user perceives to be biologically relevant in the workflow specification.

Bio
Susan B. Davidson received the B.A. degree in Mathematics from Cornell University, Ithaca, NY, in 1978, and the M.A. and Ph.D. degrees in Electrical Engineering and Computer Science from Princeton University, Princeton NJ, in 1980 and 1982. Dr. Davidson joined the University of Pennsylvania in 1982, and is now the Weiss Professor of Computer and Information Science and Deputy Dean of the School of Engineering and Applied Science. She is an ACM Fellow, a Fulbright scholar, and recently stepped down as founding co-Director of the Center for Bioinformatics at UPenn (PCBI).

Preceeding the formation of the PCBI, Dr. Davidson was involved with planning and administering an NSF funded research training program in computational biology, which has been run at the University of Pennsylvania since 1995. She also helped establish undergraduate degree programs in bioinformatics and computational biology run through the departments of Biology and Computer and Information Science, as well as tracks in this field in the Masters of Biotechnology degree program.

Dr. Davidson's research interests include database systems, database modeling, distributed systems, and bioinformatics. Within bioinformatics she is best known for her work in data integration, XML query and update technologies, and more recently provenance in workflow systems.

---