The Distributed Messaging Systems group at IBM T. J. Watson Research Center focuses on developing advanced technology for Messaging Middleware. In plain English, we are focusing on getting information to the right place at the right time.  There may be millions of pieces of information, and millions of people or computers interested in it, and very little time to deliver it.

We want to broaden the scope of messaging systems to be well-suited for development of complex distributed applications. In particular we are interested in three core problems:

• How to determine who needs what data,
• How to scale messaging systems to thousands or millions of communicating applications,
• How to simplify the development of messaging applications.

• The Gryphon project has developed a highly scalable and available middleware for the publish/subscribe messaging model. Gryphon has been deployed over the internet for real-time sports score distribution at the US Tennis Open, Ryder Cup, and Australian Open, and for monitoring and statistics reporting at the Sydney Olympics. Much of the techniques and algorithms developed for Gryphon have been incorporated into IBM Messaging products.
• The SMILE project focuses on development of a smart messaging middleware for simpler and more efficient application integration, where middleware behavior is programmed in a high level language, like SQL, and subscriptions are stateful views of published event streams. Key research is in incremental stream transforms and fault-tolerance.
• The Web Messaging project extends the reach of asynchronous messaging to Web browser-based client platforms. Web content providers have enhanced their offerings by including frequently updated data like stock quotes and sports scores. However, providers have typically used standard synchronous techniques to pull data from Web servers. Asynchronous (push) messaging fits the task more naturally, providing more timely data delivery with greater efficiency.
• We are investigating scalable distributed resource management techniques for complex message flows that consume CPU, network and disk resources, based on customer service level agreements (SLAs). Key research is in resource optimization, monitoring and modeling system behavior, monitoring SLA compliance, and resource scheduling mechanisms.