Grid yourself: No-cost code and tools from IBM
- By John K. Waters
They say that the best things in life are free. They (whoever they are) probably weren't thinking about software developers, but that saying could well apply to IBM's latest freebie: no-cost code and tools for building e-business apps for grid computing environments, which the company announced last week.
Specifically, visitors to IBM's alphaWorks Web site (www.alphaworks.ibm.com) will be able to download free trial versions of new grid technologies: the Directory Replicator for Grid and Manageability Services for Linux; new tools for Eclipse, the open-source Java development platform, including Application Tools for Extension Services and Visual Application Builder; and so-called autonomic computing resources that allow developers to create self-managing systems and infrastructures, including Generic Log Adapter for Autonomic Computing, Emerging Technologies Toolkit 1.1 and IBM Dynamic Logical Partitioning Tool Set. The free code and tools are available at no cost for 90 days.
Big Blue is betting big on the grid-computing model, and it wants to put the tools for developing grid-based enterprise applications into the hands of developers. This latest offering represents another step in the company's overall strategy to deliver on-demand computing, an architecture in which the IT infrastructure acts as a pool of resources that can respond dynamically to business demands -- essentially, a utility computing model. Researchers at Gartner have predicted that utility computing will become part of the computing mainstream by 2005.
Grid computing is the heart of IBM's on-demand computing strategy. The ability to link heterogeneous, geographically dispersed computers into a "grid" that puts idle computing cycles to work on the same computational problem simultaneously has powerful implications for budget-strapped enterprises. IBM is reportedly making all of its key software and hardware offerings compatible with grid computing.
Autonomic computing is another key component of IBM's on-demand computing model. The company coined the term in 2001 to define self-managing computing systems capable of monitoring and correcting themselves while keeping the complexity invisible to the user.
When it comes to grid computing, IBM is definitely eating its own dog food. The latest version of IBM's WebSphere app server, scheduled to ship later this month, will include a grid-computing feature. And its own Distributed Terascale facility is a grid-computing project that connects a number of machines, allowing researchers to perform 13.6 trillion calculations per second -- that's about a thousand times faster than IBM's famous chess-playing Deep Blue.
For other Programmers Report articles, please go to www.adtmag.com/article.asp?id=6265
John K. Waters is a freelance writer based in Silicon Valley. He can be reached