|
 What is new and revolutionary about ‘Computing in the Cloud’ are the entrepreneurs who are willing to bet their future to create this kind of infrastructure and package and offer computing as an on-demand service
By Satish Joshi, Patni, November 13, 2009, 1100 hrs
Ever heard of a project called SETI, Search for Extra Terrestrial Intelligence? SETI is a worldwide research effort coordinated by the SETI Institute to locate intelligence in outer space. One particularly interesting project among the many that were launched was called SETI@home. This program was designed to systematically analyze all the data gathered through astronomical observations, space telescopes like Hubble, probes sent in space and data submitted by amateur astronomers. Since nobody really knew what to look for, every conceivable piece of data had to be examined and the quantum of data was so massive that no battery of supercomputers was capable of analyzing it in any reasonable period of time. Chunks of this data and components of the analysis routines were therefore parceled out over the Internet to almost 3 million volunteers across the world. The program was written like a screen saver to automatically start whenever the PC was idle, fetch the next chunk of data from the Internet and ship the results back over the Net to the SETI Institute for collation. At any given moment any part of the program may be executing literally anywhere in the world analyzing any part of the data.
It’s a pity that perhaps the glamour associated with the search for extra terrestrial intelligence so completely overshadowed the technological achievements of SETI@home. It’s also a pity that the developers of this technological marvel did not exploit their creation commercially. Had they done so, today, they’d have been known as the visionaries who created Cloud Computing!
Fast forward to the 90s and we come across an operating system (OS) that IBM called VM i.e. the Virtual Machine. It was a platform to run other OS but one could simultaneously create multiple virtual machines running perhaps CMS (another OS) or OS/VS1 or DOS/VSE and run applications written for that OS in the relevant virtual machine. VM would manage the overall computing load and demand for computing resources, dynamically allocating them to various virtual machines to maximize the overall utilization and throughput. The users could shift or distribute their computing tasks to such VM as required.
And then there was the term ‘Distributed Computing’? It’s really a simple concept where a large computing problem gets divided into smaller simpler problems. These smaller problems get distributed to different computing systems and get solved more or less independently of each other. The final solution gets constructed from the solutions obtained from each of the simpler computations. The objective: to do a super computing task without really buying a super computer. However, it never became known to those whose problems it could have actually solved and who would have understood what it meant.
What’s all the hype about? >>
l Page 2 l Page 3 l
|