Moore's law has driven CPU power and memory capacity to grow million times since the system R and Ingres projects started thirty years ago. The underlying software technology has also changed substantially. Today, operating systems support virtually infinite address and lightweight multithread library for efficient utilization of multiprocessor systems with very large memory.

Despite these dramatic advances in underlying hardware and software, the fundamental DBMS architecture remains largely unchanged. Data and indexes are managed as disk-resident structures. The heavyweight process architecture which involves costly context switch overhead is still dominant. Typical commercial RDBMS implementations involve several millions of lines of complex code that has been evolving over decades. Because it is extremely risky to overhaul any software of this size, commercial RDBMS implementations are likely to maintain the current, disk-resident, heavyweight-process architecture.

On the application side, there is growing demand for real-time acquisition and analysis of a large volume of data, especially, continuously arriving stream data. Examples are traditionally found in financial services, telecom, defense and intelligence, logistics, scientific discovery, and this list is being expanded to include other domains such as supply chain and retail with the development of RFID technology for ubiquitous tracking of physical objects. These real-time enterprise applications demand orders of magnitude higher performance scalability than existing RDBMS implementations can manage.

Recent research addressed the growing impedance mismatch between DBMS implementations and modern hardware architecture. For example, latch coupling, a common technique for concurrency control of shared index nodes incurs excessive coherence L2 cache misses on shared-memory multiprocessors. L2 cache misses are expensive, costing the loss of a thousand instruction opportunity as the gap between CPU and memory speeds grows. Other types of L2 cache misses occur because disk-resident data and index structures are not optimally laid out for CPU processing. With this impedance mismatch, there is a limit in improving the database performance by just increasing the hardware capability.

To exploit the processing power of modern hardware architecture and to meet the database performance demands of challenging applications, it is necessary to reexamine the underlying premises of the current DBMS architecture from the new perspective.

The objective of this panel is to promote the research on new DBMS architectures within the database community. The panelists will review the current state of research, discuss market needs and requirements, and explore future research directions.

Sang K. Cha
Sang K. Cha has worked on memory-centric database architecture since he was involved in an in-memory query processing project at HP Laboratories in 1991. In 1992, he joined Seoul National University and initiated his own lab to further investigate memory-centric database architectures. This effort led his team to develop two generations of memory-centric database engines over a decade. The first one based on conventional architecture became the basis of numerous in-memory database systems in production at Korean telecom and financial institutions. In 2000, he founded Transact In Memory with his students to develop a new incarnation overcoming the limitations of the first-generation memory-centric database systems. Based on the successful deployment of this new system in Korea, his team is working with global software and hardware vendors to solve the real-world problems with challenging scalability. He received his Ph.D. from Stanford University, and his MS and BS from Seoul National University. In 1980's, he was an early employee of DACOM, the first private Korean telco which set the cornerstones for Korea to develop today's advanced IT infrastructure. He also worked at AI Lab of Texas Instruments, Inc.

Anastassia Ailamaki
Anastassia Ailamaki received a B.Sc. degree in Computer Engineering from the Polytechnic School of the University of Patra, Greece, M.Sc. degrees from the Technical University of Crete, Greece and from the University of Rochester, NY, and a Ph.D. degree in Computer Science from the University of Wisconsin-Madison. In 2001, she joined the Computer Science Department at Carnegie Mellon University as an Assistant Professor. Her research interests are in the broad area of database systems and applications, with emphasis on database system behavior on modern processor hardware and disks. Her projects at Carnegie Mellon (including Staged Database Systems, Cache-Resident Data Bases, the Fates Storage Manager, and PUMA2), aim at building systems to strengthen the interaction between the database software and the underlying hardware and I/O devices. Her other research interests include automated database design for scientific databases, storage device modeling, and internet querying. She has received three best-paper awards (VLDB 2001, Performance 2002, and ICDE 2004), an NSF CAREER award (2002), and IBM Faculty Partnership awards in 2001, 2002, and 2003. She is a member of IEEE and ACM.

Yoshinori Hara
Dr. Yoshinori Hara serves as Chief Research Manager, NEC Corporation, and is currently in charge of NEC Kansai Research Laboratories located in Japan. He is responsible for conducting research and development on advanced ubiquitous computing including Web/Hypermedia systems, mobile & embedded systems, adaptive user interfaces, advanced information retrieval technologies, system security & reliable systems, etc. Since he joined NEC Corporation, he has been serving various research positions in NEC R&D Organization. Prior to the current position, he worked on establishing a new research organization in the Silicon Valley in 1995 and was serving as Department Head, NEC Laboratories America, Inc. From 1990 to 1991, he was a Visiting Researcher at the Department of Computer Science, Stanford University. He received his B.E. and M.E. from University of Tokyo, and his Ph.D. from Kyoto University.

Wei Hong
Dr. Wei Hong is a senior researcher and principal investigator at Intel Research Berkeley where he leads the sensor network project. His current research focuses on data management issues in sensor networks. He co-authored the Best Paper of VLDB 2004. He co-architected and developed TinyDB, a widely used, open-source, in-network distributed sensor database system deployed in real-world applications. Prior to joining Intel Research, Dr. Hong co-founded and architected two startup companies: Illustra Information Technology Inc. and Cohera Corp. Illustra developed the first successful commercial Object-Relational database system. It was acquired by Informix, now part of IBM. Cohera provided electronic catalog management solutions based on a novel federated database system that it developed. Its technology was acquired by PeopleSoft. Dr. Hong earned a Ph.D. in computer science from UC Berkeley and holds a master and two bachelor degrees from Tsinghua University in Beijing, China.

Martin Kersten
Kersten received his PhD in Computer Science from the Vrije Universiteit in 1985 on research in database security, whereafter he moved to CWI to established the Database Research Group. In his professional carreer, he has developed three complete database kernels.
From 1979 until 1985 he developed a small relational kernel, called Troll, which was sold as part of a CASE tool 1985-1991. Between 1986 and 1991 he was co-designer of the PRISMA database machine, a RDBMS for a 100-node multiprocessor based on the assumption that the hotset is memory resident. In 1992 he initiated the development of his 3rd DBMS, called MonetDB. This system is an extensible main-memory oriented DBMS which is currently used in a commercial data mining system and the pivot in several national projects aimed at advanced database applications, such as image processing and geographical information systems.
Currently he is heading a department involving 50 researchers in areas covering datamining and datawarehousing, multimedia information systems, information engineering, and quantum computing. Since 1994 he is professor at the University of Amsterdam. In 1995 he co-founded Data Distilleries, an SME specialized in data mining technology.
He has published ca. 170 scientific papers. He acts as a reviewer for ESPRIT projects and is a trustee emeritis of the VLDB Endowment board.

Tore Risch
Tore Risch is Professor of Database Technology at Uppsala University (Sweden) where he leads the Uppsala DataBase Laboratory (UDBL) research group (http://user.it.uu.se/~udbl). He was previously Professor at Linköping University (Sweden). Before Linköping he was staff member in the Database Technology Department at Hewlett-Packard Laboratories (Palo Alto, California), and Visiting Scholar from HP in the database group at Stanford University. Prior to joining HP, he worked for Syntelligence Inc. (Sunnyvale, California) designing a product for large scale knowledge bases combining AI and database technologies. He also worked on the Prospector expert system (SRI, Menlo Park, California), on integrating Prolog with relational databases (Uppsala U., Sweden), and at IBM's Almaden Center (San Jose, California) on functional knowledge representation. He made his PhD 1978 at Linköping University (Sweden), on query optimization in a meta-database system. He is author of over 100 technical publications and 4 US patents. He was General Chair of two international scientific conferences and member of the program committees of 42 international scientific conferences.

Vishal Sikka
Dr. Vishal Sikka is Vice President of Advanced Technology at SAP where he works on new technologies in enterprise infrastructure (SAP NetWeaver) and applications. Prior to joining SAP, Vishal was area Vice President of platform technologies at Peregrine Systems (Remedy) responsible for Peregrine's efforts in application development and data integration. Vishal joined Peregrine following the acquisition of Bodha, Inc. where he served as founder and CEO. Bodha developed technology for non-invasive, service based integration of enterprise applications and information.
Vishal holds a Ph.D. degree in Computer Science from Stanford University. His experience includes research in automatic programming, information & application integration, and artificial intelligence at Stanford, Xerox Palo Alto labs, and two startups.

Business Process Intelligence (BPI) refers to the application of business intelligence (OLAP, data mining, analytics, etc) techniques in business process management, with the goal of providing a better understanding of a company's processes and of devising ways to improve them. It is an area that is quickly gaining interest and importance in the industry. BPI includes many sub-areas. As indicative examples, we list the following:

• Process discovery: this refers to the analysis of enterprise operations in order to derive the process models that these operations obey.
• 'Intelligent' process analysis: this refers to the analysis of business process execution to discover interesting correlations, e.g., between process data and resources and business metrics, to perform capacity planning, or to identify the causes of low-quality process executions.
• Prediction: besides analyzing the value of business metrics and understanding, among other things, the causes of low-quality process executions, BPI aims at predicting critical situations (e.g., an exception, or a delay) on a running process instance before it actually happens.
• Exception handling: once a problem has been recognized (or predicted), another goal of BPI is to assist the analyst in making decisions to address the problem.
• Static optimization: the intelligent analysis described above may lead to the identification of areas of optimization for a process, for example in terms of different sizing of resource pools, different resource assignment criteria, and the like. BPI offers support for optimizing the process configuration to improve upon those areas.
• Dynamic optimization: ideally, one could think of an intelligent component that constantly manages and supervises each process instance (in a controlled way), for example by having influence in routing and task assignment decisions in order to maximize certain business objectives.

The panel will involve lively, controversial discussion among leading experts in the field, from both industry and academia. In fact, we believe that the topic is very interesting from both an industrial perspective (clear business potential and need from companies) and an academic perspective (tons of interesting and challenging problems, most of which are very, very far from being solved and even understood). We invite every ICDE attendee to join the panel and participate in the lively discussions to try to identify the hardest and more interesting issues in this novel research field.

Malu Castellanos
Malu Castellanos has been working at HP Labs for the last 7 years, where as a senior researcher she has been developing and applying data and text mining techniques to different kinds of problems and solutions, including the analytical functionalities of the BPI platform. Previously she was an Associate Professor in the Information Systems Department at the Polytechnic University of Catalunya. She has more than 30 papers in international conferences and journals and has served as chair and PC member for a number of conferences. She is coauthor of a survey book on text mining and a book on multidatabase systems. Her research interests are data mining, text mining, databases, business processes, and interoperability. She received a degree in Computer Engineering from the National University of Mexico, and a Ph.D. from the Polytechnic University of Catalunya (Barcelona).

Fabio Casati
Fabio Casati is a senior researcher at HP Labs, Palo Alto. He got his PhD from Politecnico di Milano (Italy) in 1999. His research interests include business processes, Web services, business-aware application management, and "middleware intelligence" (embedding data mining technologies into the middleware). He has led the development of several applications and is author of more than 60 papers in international conferences and journals. He initiated the research in Business Process Intelligence and organized the first conferences and journal issues on e-services. He is also co-author of a book on Web services. Fabio has also served as chair and PC member for dozens of conferences in the areas of databases, information systems, and Web services.

Scott Fingerhut
Scott Fingerhut is the General Manager of TIBCO's Business Optimization product line, which includes the company's portal and business activity monitoring (BAM) products. He has published articles on application integration, business intelligence and data mining and has spoken at a variety of industry and government conferences. Fingerhut previously held marketing and public relations management positions at the world's largest global public relations agency, Weber-Shandwick, representing customers that included Compaq and Hewlett-Packard. He holds a Master's degree in organizational management and a Bachelor's in business management from San Diego State University.

Ramesh Jain
Dr. Ramesh Jain is a Donald Bren Professor in Information & Computer Sciences at Bren School of Information and Computer Sciences at University of California, Irvine. Previously Farmer Chair and GRA Eminent Scholar at the School of ECE and College of Computing at Georgia Tech, Atlanta. He was the founder or co-founder of a number of companies, the last one being PRAJA inc, which was a leading provider of infrastructure software technology for development of event-based information management for experiential systems with customers like Yahoo and GM. It was acquired by Tibco Inc. He has been an active researcher in multimedia information systems, image databases, machine vision, and intelligent systems. While professor of computer science and engineering at the University of Michigan, Ann Arbor and the University of California, San Diego, he founded and directed artificial intelligence and visual computing labs. He was also the founding Editor-in-Chief of IEEE MultiMedia magazine and Machine Vision and Applications journal and serves on the editorial boards of several magazines in multimedia, business and image and vision processing. He has co-authored more than 250 research papers and books. Dr. Jain has been elected Fellow of ACM, IEEE, IAPR, AAAI, and SPIE.

Guy M. Lohman
Dr. Guy M. Lohman is Manager of Advanced Optimization in the Advanced Database Solutions Department at IBM Research Division's Almaden Research Center in San Jose, California, and has 22 years of experience in relational query optimization. He is the architect of the Optimizer of the DB2 Universal Data Base (UDB) for Linux, Unix, and Windows, and was responsible for its development in Versions 2 and 5, as well as the invention and prototyping of Visual Explain. During that period, Dr. Lohman also managed the overall effort to incorporate into the DB2 UDB product the Starburst compiler technology that was prototyped at the Almaden Research Center. More recently, he was a co-inventor and designer of the DB2 Index Advisor (now called Design Advisor), and co-founder of the DB2 Autonomic Computing project (formerly known as SMART -- Self-Managing And Resource Tuning), part of IBM's company-wide Autonomic Computing initiative. In 2002, Dr. Lohman was elected to the IBM Academy of Technology. His current research interests involve query optimization and self-managing database systems.