IBM Research was founded in 1945 and has since generated a string of technology innovations that have had a tremendous impact on IT, including Fortran, reduced instruction set computer (RISC) architectures, relational databases, token-ring networking and magnetic disk storage. Following major reductions in budget and staff during the mid-1990s, IBM Research is again on a growth path and employs approximately 3,000 researchers in eight research centers worldwide.

Research Areas: IBM Research engages in an extremely diverse range of research that cuts across geographic boundaries. Notable research themes include:

• Dynamic e-business — aims to integrate business processes and leverage the emerging e-business infrastructure of Web services, standards and messaging to automate internal and external operational decision making, transactions, collaboration and execution with business partners
• Deep computing — builds on the victory of IBM’s Deep Blue chess computer over world champion Garry Kasparov in 1997 to address computationally challenging problems such as weather prediction and gene sequencing
• Mining of unstructured data — allows text, voice and XML information captured from diverse sources (e.g., customer calls, meeting discussions, company intranets, the Internet) to be managed and analyzed in new ways
• Autonomic computing and its IBM "eLiza" implementations — seeks to change how large-scale servers, network storage, software platforms and eUtilities are built and managed
• Advanced user interfaces — includes speech recognition and gaze tracking
• Pervasive and mobile computing, in particular the middleware and architectures — supports mobile, embedded and wearable devices
• Basic science — includes chip technology, nanotechnology, computational biology and materials science

The research area most heavily hit by staff reductions has been the basic sciences. Despite this loss, IBM remains one of the few industrial R&D labs (along with Bell Labs) to conduct substantial research into physics and materials science. This basic research has led to successes in areas such as semiconductors (e.g., high-performance, low-power silicon germanium chips and copper chips) and storage (e.g., the one-inch-square, 1 GB microdrive and higher-density "pixie dust" disks).

Technology Transfer: In core computing technologies, IBM follows a technology transfer path of gradually adding development and engineering staff to promising projects, eventually forming new product teams. This approach works well with technologies with a long, high-cost development cycle, such as hardware, semiconductors and “deep” areas of software, which make up approximately 20 percent of IBM Research’s efforts (compared to about 80 percent 30 years ago).

For research with a potentially faster route to commercialization, IBM Research has evolved alternative mechanisms to demonstrate and promote the feasibility of new approaches. In particular, many technologies require a pilot phase where IBM Research works directly with end-user organizations. For these situations, IBM uses three approaches to technology transfer.

1. Focus on Major Clients: As a special service for IBM’s largest clients (usually through outsourcing deals), IBM examines other client needs that match with activities in Research and works jointly with the client to help direct the research (e.g., pattern discovery for a pharmaceutical company).

2. First of a Kind: Once IBM Research feels a technology has reached a level where practical benefit can be achieved, it partners with a leading-edge client that is prepared to try the technology in a real-world situation. One early example of a highly successful "first of a kind" project was IBM's teaming with New York’s Memorial Sloan-Kettering Cancer Center and Massachusetts General Hospital in Boston. The project resulted in MedSpeak, a specialized speech recognition application for radiologists, whose distinct technical vocabulary made recognition easier. As the technology improved, IBM expanded into legal dictation and then general products, establishing its ViaVoice line as a leader in the speech recognition marketplace.

3. Emerging Business Opportunities: Following a practical demonstration through a first-of-a-kind project or similar initiative, IBM works with a broader range of clients to examine the common needs for the technology and identify strategies to bring the technology to market. Technologies at the emerging-business-project phase include:

• "Project WF" — looks for relationships in unstructured information and uses crawler technology to bring back terms and relationships (e.g., learns from statistical analysis that Lou Gerstner is a person, and is related to IBM through the “chairman of the board” relationship), with the aim of improving access to and management of unstructured information
• Distributed matchmaking — uses innovative algorithms for complex matching of buyers and sellers in e-marketplaces
• High-volume messaging — examines the future of messaging, including publishing large amounts of dynamically changing content while handling very large peak loads, building on IBM’s experience in managing highly accessed Web sites such as Wimbledon and the U.S. Open tennis tournaments

Strengths and Challenges: IBM Research is one of the few remaining commercial U.S. research laboratories conducting pure research, which leads to a culture of academic excellence, openness and collaboration. Its strengths lie in the breadth and depth of its research and its ability to attract and retain world-class researchers. It actively manages the relationship between pure and applied research to ensure continuing relevance to IBM’s products. IBM Research is proactive in creating and evaluating technology transfer programs, and IBM’s consulting organization is an additional asset in forging links to real-world customers.

With such a broad range of technologies under development, one of the challenges for IBM Research is how to communicate its activities internally and externally without overwhelming the audience. Research pays attention to this issue by producing professional journals and publications and maintaining a well-organized, informative and up-to-date Web site. Another issue arises from IBM’s role as one of the most aggressive companies in bringing new technologies to market, which leads to inevitable failures that may confuse the marketplace as products are abandoned (e.g., early forays into signature verification, and the more-recent marketing of the CrossPad handwriting capture device).

Strategy: To balance fundamental and applied research aligned with IBM's product and customer needs.

Strengths:

• Breadth and depth of research
• Facilities, culture and research attract world-class researchers
• Proactive technology transfer
• Aggressive introduction of innovation into products

Challenges:

• Maintaining focus in driving, managing and communicating broad research activities
• Managing unsuccessful products or innovations

Bottom Line: Because of the breadth and depth of its research, focus on technology transfer and communication, and willingness to innovate in the marketplace, IBM maintains its lead as the most influential IT research laboratory in the industry. Its challenge will be to continue current levels of activity and innovation in the face of potential cutbacks by customers in an economic downturn. IBM Research (along with IBM Global Services) should be considered by enterprises needing an IT partnership that requires a significant direct IT research engagement.

This research is part of a set of related research pieces. See AV-14-7880 for an overview.

Entire contents (C) 2001 by Gartner Group, Inc. All rights reserved. Reproduction of this publication in any form without prior written permission is forbidden. The information contained herein has been obtained from sources believed to be reliable. Gartner Group disclaims all warranties as to the accuracy, completeness or adequacy of such information. Gartner Group shall have no liability for errors, omissions or inadequacies in the information contained herein or for interpretations thereof. The reader assumes sole responsibility for the selection of these materials to achieve its intended results. The opinions expressed herein are subject to change without notice.