IBM unveils 'self-healing' tools

ARMONK, N.Y. - IBM moved closer to achieving its vision of self-healing, self-configuring systems last week with the unveiling of three autonomic computing software modules.

The modules are designed to predict and respond to sudden increases in data center workloads. The Adaptive Forecasting module uses mathematical models to anticipate the progression of an unexpected surge in demand. The Online Capacity Planning module estimates the resources required to maintain service-level targets during peaks and allows a hot swap of resources from one workload to another. The Rapid Reconfiguration piece uses new capabilities in WebSphere Application Server 5.0 to add and remove nodes as resource demands fluctuate.

While IBM has announced dynamic provisioning tools in the past, what makes this software different is its forecasting abilities, says Alan Ganek, vice president of autonomic computing. The software will start to deploy equipment ahead of data center requirements.

"Obviously, you can't tell in advance when a surge is going to happen," Ganek says. "But when the workload starts to come in, we can gauge a signature of what it looks like. And using modeling and probabilistic methods, we can get a sense of its trajectory."

Click to see:

IBM has existing forecasting tools that can prepare for predictable spikes, such as seasonal traffic bursts. But dealing with the unexpected requires something more sophisticated, says Joe Hellerstein, manager of the adaptive systems department at IBM's Thomas J. Watson Research Center.

When workloads deviate significantly from what is expected, IBM's new adaptive forecaster kicks in and starts tracking how things are changing. This short-term forecaster allows the system to get ahead of the surge, see that more server power will be needed in another minute, for example, and make requests before service levels degrade, Hellerstein says. Similarly, as the surge begins to dissipate, the tools can release resources earlier to conserve costs, he says.