A couple weeks ago I linked to a Dan Farber article on 3Tera's AppLogic grid processing application-a scalable, on-demand, commercially available processing platform for businesses that want to leverage the power of massively parallel processing on commodity hardware, but don't want to start from the ground up building or customizing an operating system which will do so.

What I said in my post was that it sounded a lot like what Google had already done, and that being the case, why weren't they commercializing it in a similar fashion? I've considered Google's back-end their secret weapon for a long time now-that's the real success story at Google, and it provides them with an amazing amount of flexibility.

Bert Armijo of 3Tera was kind enough to respond with his perspective, which I will re-post and respond to below the jump.

First, Bert acknowledges that Google's developments are substantially similar.

Yes, Google's done it. In fact, they consider their operations systems part of their core competence and are very tight lipped about it. What we at 3tera are doing is offering that same capability to everyone who operates an online service.

Fair enough-Google's secrecy about their back end is well-known, and no one I know has even heard hints that they might be opening it up. Yet the power and utility are obvious; so it makes sense that someone (3tera in this case) would want to commercialize such a system and that prospective buyers who understood the advantages might abound.

But the real question, then, is if the power and utility are obvious, and the market exists, then why isn't Google getting into it? Do they not see the opportunity, do they have a better one to focus on, or is there really no opportunity at all? Or are they simply biding their time and plotting their own move in secrecy? To be fair, this a question for Google (and we all know Google doesn't like to answer questions) not 3tera, but Bert takes a swing at it as well.

This isn't a unique phenomenon.

Lots of folks built operating systems before Microsoft and Linus, but most of them were large companies selling computers and saw the OS as a tool rather than a product. As such, they built something specific to that task. More recently, the big players didn't jump into virtualization until VMware had proven the market because at the time they'd rather you just bought another server and/or another license.

Back to utility computing, consider Amazon's EC2. They've been pretty open about the fact that it was built to run Amazon and that's evident in it's feature set. If you started building a commercial utility computing system from scratch you'd probably consider it a given that it must run databases, have predictable performance, and allow installing code directly on the system - but EC2 provides none of these. I'm not trying to disparage Amazon, I think they've done a pretty good job, but if you read their forums you'll see their customers have been asking these questions since the service first came out more than a year ago.

More advanced requirements would never be needed, or even desired, by larger players. For instance we work with multiple hosting providers in several countries to give users a choice of data centers. To make it possible to use multiple providers or even switch, AppLogic allows full portability of applications (no matter how complex) with a single command.

This is a fair assessment; I'm not sure the OS metaphor is entirely apt, however, and if not it breaks the logic of the argument. It's true that many operating systems were designed for specific tasks and thus poorly suited to widespread distribution. And there are many types of grid computing that also fit that bill; as Armijo mentions, Amazon's EC2 among them-it was clearly designed to run Amazon's store backend, and hasn't proven well-suited to all other applications.

But Google's development seems as though it may be a different beast; they haven't had any apparent trouble adopting it to a wide variety of disparate tasks. It may have been designed for search at the outset, but it's done quite well supporting mail, web-based office suites, and any number of other application acquisitions they have made recently. So I think it is hard to say that it suffers from a specificity issue in the same way that EC2 does. Of course, it may have other problems stemming from its origins, it's hard to say given what we know (which I think is part of Armijo's point) whereas 3tera's application was designed from the ground up with the sort of utility and features that a commercial product destined for broad markets ought to include.

From the perspective of a CIO in the market for such grid computing power, 3tera's hosted solution plan seems like the only realistic choice at the moment (I do like that AppLogic is also available as a stand-alone product which can be purchased to build out your own grid with-I wouldn't generally recommend doing so, but I like the options) given some of the limitations shown by EC2. Still, if times weren't pressing, I might hold back and see what happens in this marketplace. Google has other problems and if, as Bert and I agree, their backend is their core competency, there may come a day when they turn to it for leverage with a non-commodity product for their portfolio.