GRIDtoday
The Leading Source for Global News and Information from the evolving Grid ecosystem,
including Grid, SOA, Virtualization, Storage, Networking and Service-Oriented IT
February 26, 2007
From the Editor
Defining Grid: Who Cares?

Fresh off a brief sabbatical, I have to say it's good to be back in the Grid world. After all, where else can a guy lose himself in a debate over semantics while the world around him is simultaneously adopting and advancing -- without regard for what it's called -- that very technology?

Case in point: I recently read a piece from eChannelLine discussing The 451 Group's January report on the state of the Grid market. In recapping the report's contents, the author focuses on the debate over terminology, where the past year has seen the term "Grid" fall out of favor and get replaced by a host of other -- in my opinion -- equally confusing, vague terms. The author cites 451's Steve Wallage as such: "Terminology is a major challenge in this area. More than 70 percent of our base of grid users preferred a term other than grid to describe their deployments, although there was no common agreement as to what this term should be." Wallage went on to state, however, that Grid often is at the core of companies' next-generation datacenter, virtualization and service- oriented IT strategies.

No more than 20 minutes later, I was reading a blog entry by Mark Cuban, renegade owner of the NBA's Dallas Mavericks, about "The Future of Personal Computing." Cuban, who has been and continues to be involved with business ventures in the high-tech realm, sees a future where home users abandon their high-end PCs and expensive software for thin-client consoles (maybe even next- generation gaming consoles) and free apps online, all powered by Google and its massive datacenters. Writes Cuban:

"Spreadsheets, Word Processors, Powerpoint, Email, Personal Database, Personal Finance? They are all so limited in their resource utilization, a very strong case could be made that its [sic] smarter to do these things online from a gaming console or any inexpensive PC. The inherit benefits of distributed computing could outway [sic] some of the limits of not having the biggest box on the block."
... Huge datacenters of clustered computers running an unlimited number of Virtual Machines for an unlimited number of users with unlimited bandwidth out, all free to consumers in exchange for seeing ads in limited areas could turn the consumer world upside down."

So, what does this all have to do with Grid computing, which doesn't appear by name in any of the trends Wallage cited or in Cuban's distributed vision of the future? Well, as Wallage pointed out, the principles of Grid computing -- in its infinite broadness -- comprise the core of these technologies. The lesson we can take from this, in my humble opinion, is that regardless what vendors and standards bodies call their technologies, and regardless how many reports and conference sessions try to drill down to the true meaning of the word "Grid," most people just don't care. They know the underlying technology is effective, and that's all that matters. With this in mind, my advice would be to call it what you will -- Grid, virtualization, distributed computing, utility computing, fabric, etc. -- and forget about it. Focus your energy on the technology instead of the terminology.

And, lest you're thinking of abandoning the word "Grid" because the Web 2.0 generation doesn't know the meaning, I offer you this little gem from the blogosphere on how to build a small-scale YouTube using, you guessed it, Grid computing. Maybe the word isn't destined for the glue factory, after all.

As for this week's issue, we're giving some love to the big, scientific production grids that made everything possible. Our lead article, "Keeping Pace With the Future: The History and Future of Itanium-Based Systems on the Grid," discusses how Intel's Itanium processor revolutionized the world of high-performance Grid computing in projects like TeraGrid and CERN's LCG, as well as how members of the Gelato Federation look to further Itanium's role in the world of research-driven Grid computing. All in all, it's a very well-done article that speaks volumes about the ubiquity of Grid computing, which is being done by highly funded international consortiums and small, foreign universities alike.

-----

Comments about GRIDtoday are welcomed and encouraged. Write to me, Derrick Harris, at editor@gridtoday.com.