IBM: We're Making Strides With Autonomic Computing

05/19/2004

IBM says three IBM Business Partners will integrate autonomic computing technologies into upcoming products available this month. Additional business partners have committed to release products later this year using IBM autonomic computing technologies. IBM also announced key research projects in progress designed to further drive innovation in autonomic computing.

IBM offered these details:

Corente, NetFuel Inc. and Singlestep Technologies are scheduled to introduce products this month with the Common Base Event format, previously submitted by IBM to the OASIS standards body, which is envisioned as the basis for standardized exchange of problem determination data. The companies also integrated the Autonomic Management Engine (AME) into upcoming products. The AME monitors events, analyzes them, then plans and executes corrective action on a computing resource. When integrated with the other autonomic technologies, the AME is the facilitator of a self-managing system.

Addamark Technologies and Network Physics have declared they will debut products later this year with IBM autonomic capabilities.

A few weeks ago, IBM and Cisco jointly announced the Cisco Business Ready Data Center Optimized with IBM for the on demand operating environment. Autonomic computing technology was implemented for the log conversion of Cisco's IGESM in the BladeCenter, which will provide customers the unique capability to pinpoint and resolve potential networking related problems using autonomic problem determination technologies.

In 2001, IBM Research's Paul Horn introduced the term "autonomic computing" and published the autonomic computing manifesto, a call to action to the industry and an overview of autonomic computing systems. IBM has since launched the Autonomic Computing unit; developed an industry blueprint for autonomic computing; integrated autonomic computing capabilities into more than 415 product features in 50 distinct IBM products; continues to drive the adoption of standards across the industry; and provides assets and tools to assist software developers in designing and testing autonomic solutions.

IBM also works with the academic and research communities to further the discovery and development of autonomic computing technology and research. IBM recently announced Georgia Institute of Technology's plans to use IBM hardware and software to advance the university's development of technology for autonomic computing.

"We have made significant progress with incorporating autonomic computing capabilities into IBM products," said Alan Ganek, vice president, Autonomic Computing, IBM. "We are very excited to see the increasing interest from the industry to embrace autonomic computing -- from IBM Business Partners to customers to the academia community. As we continue in our drive to create self-managing systems, the innovation from IBM Research will be critical in the evolution of autonomic computing."

IBM Research continues to make great strides to further the vision of Autonomic Computing, a world in which computing systems manage themselves to a far greater extent than they do today. It is a world, in particular where interacting sets of individual computing elements regulate and adapt their own behavior in order to respond to a wide range of changing conditions with only high-level direction from humans.

IBM Research is focusing its efforts on developing technologies and architectures that support multi-component self-managing systems. Large-scale systems, composed of interacting assemblies of these components and others that manage and serve them, will be self-configuring, self-optimizing, self-healing, and self-protecting. Key research initiatives include:

--Unity: The Unity project explores component behaviors, relationships and technologies that will support self-management of complex computing systems. Whether a component is a database, storage system or server, each element is responsible for its own internal autonomic behavior, including managing the resources that it controls, and managing its own internal operations. Each element is also responsible for forming and managing the relationships that it enters into with other autonomic elements in order to accomplish its goals. The Unity project has created a working prototype of an autonomic data center that configures, optimizes and heals itself.

--Change Management with Planning and Scheduling: CHAMPS: Change management is a process by which IT systems are modified to accommodate considerations such as software fixes, hardware upgrades and performance enhancements. Today, administrators perform these tasks with little automation for planning and scheduling changes, either off-line or in real time. The CHAMPS system, a prototype under development at IBM Research, is able to plan and schedule change management actions automatically and in real time by exploiting knowledge of service dependencies in distributed systems. In essence, CHAMPS is able to figure out what configuration changes need to be done, plan how these changes should be rolled out, and make them happen.

--Planning for Orchestrated Remediation of Security Incidents: Elix0r: While there is considerable focus in the IT industry on increasing the accuracy and automation of security incident detection, the response and recovery procedures for these incidents tend to be manual and somewhat ad-hoc. The Elix0r project aims to build a policy-enabled system for effecting security incident response and recovery procedures to achieve graceful degradation and reinstatement of services. Security incidents range from non-compliance with a security policy to vulnerability and infection by worms and viruses.

--Event Mining: Self-healing systems should acquire the knowledge needed and adapt and grow such knowledge over time. One key type of knowledge is 'problem signatures' that indicate the underlying cause of a malfunction or slow-down. The Event Mining (EM) project aims to mine historical data to construct signatures for problem diagnosis and resolution. The EM system will improve the quality and speed of signature construction by mining current data sets, the historical data generated by a distributed computer system (log files and event logs, for example), and finding correlations between events and run-time behavior. Ultimately, by finding these patterns, the cost of problem determination and resolution will be greatly reduced.

Home
Webinars
Blogs
ebizQ Forum
All Topics
Subscriptions
White Papers
Analyst Corner
Follow us on Twitter
Fan us on Facebook

Explore Our Topics

BPM Strategy
BPM Solutions
Event Processing Strategy
Content-Centric & Collaborative BPM
Process-Centric BPM

About Us
Editorial Submissions
Feedback
FAQ
Advertise With Us
Site Map
Privacy Policy
Unsubscribe