The Red Shift, best known in the field of astronomy as the measure of an expanding universe, has been adopted by Sun as a measure of growth in IT. Mike Douglas, Sun VP for global field marketing, explained to Pallavi Rao with The Economic Times, "...growth in IT is less from computing and more from social networking." To accommodate this evolving market, Sun has applied the red shift.

Sun sees certain companies as consumers of inordinate and exponentially growing amounts of IT infrastructure. Sun calls this constellation of clients a "redshift" market. These clients experience unpredictable demand for compute power and network bandwidth because their compute cycle use depends entirely upon communities using networked applications and services. Douglas reports that the demand is such that Sun has spent close to $3 million on these high-growth, network intensive applications while allocating only a third of that amount to traditional computing. "In ten years we expect 75% of our revenues to be generated by these applications," he says.

Rao cites Amazon.com as a typical example of a redshift company, where its on-demand storage service, Amazon S3, has grown to five billion objects from zero in less than a year. Another company, 3tera, has been experiencing 100% quarterly growth for its utility computing service, which easily qualifies it for membership in the redshift club, Rao writes.

Numerous other examples cited by Rao are familiar names in the Web 2.0 world. All are experiencing exponential growth in the demand for their services. Google, for example, says Douglas, "has built up that kind of capacity (the company hosts over 5 million servers) because it does not know who, where and when is going to download what data."

In Rao's view, redshift presents complex dilemmas for large companies growing faster than Moore's Law which requires better utilization of IT infrastructure, datacenter space and power. Maintaining racks of servers in conventional datacenters is stretching their IT resources beyond capacity and is not truly a solution to the problem. "The challenge therefore is designing really efficient systems to handle these kinds of workloads, and getting productive with the datacenter software and its management," says Douglas.

See Rao's article Red Shift Theory for more. [...read more...]

From the latest issue:

Recent Blog Entries as of March 18, 2010, 11:04 am
Suits and Their Countersuits A More Direct Channel Model Planned for Oracle-Sun Oracle, OpenSolaris, and Solaris Flash Storage in Exadata V2 Cuts Query Processing Time Significantly Document Freedom Day Set for March 31st	PSA Deploys GlassFish in the Manufacture of Peugeot, Citroën Vehicles Running VirtualBox VMs as Services in Windows 'Solaris 10 ZFS Essentials' Mobile Desktop Grid Oracle Management Connector for Ops Center
Blog RSS Feed