March 30th, 2005
Software utility providers prove bearish on hardware utilities
In another blog entry, I interviewed Bloglines founder Mark Fletcher. As a result of AskJeeve’s acquisition of Bloglines, Fletcher is now the general manager/vice president of "Bloglines at AskJeeves" (a mouthful that needs to be fixed). Given that Bloglines is basically a software utility, I asked Fletcher the same question that I’ve started to ask the providers of other software utilities (e.g.: salesforce.com). The idea is to do an informal survey that reality checks the idea of an on-demand hardware utility with an audience already predisposed to the idea of utility computing. I asked him whether, in addressing the scaling challenges that a software utility like Bloglines must address, the idea of a $1/CPU/hour compute utility (along the lines that Sun is going) is interesting to him at all in terms of a way to cost-efficiently scale his business.
Said Fletcher, "I’ve seen Sun’s ad. They run a billboard on 101 (in the Bay Area). We take a different philosophy. We look for the cheapest Intel hardware we can find and run Linux on it. In building this, we assume that we’re going to have failures and we design the ’system’ to tolerate them. It’s much the same philosophy that Google takes. You get so much bang for the buck that it’s a no-brainer. Also, they’re focusing on a different market. We’re not CPU bound. We’re storage bound."
As it turns out, Sun is making storage available on a $1/gigabyte/month utility basis as well. (Perhaps an awareness problem for Sun?) But its storage utility is probably best suited to customers of the current CPU utility initiative and that’s largely restricted to high performance computing (HPC) applications (that are well suited to grids). This application restriction, as best as I can tell from Sun president/COO Jonathan Schwartz’s various pitches for utility computing, is supposed to be temporary. Which is why, in a recent column, I took a more strategic look at the traction that Sun’s $1/CPU/hour message is getting.
Tactically speaking, the HPC market, where specialized CPU-bound applications like Monte Carlo simulations can be spread across a grid of low-cost systems, makes the most sense for Sun to go after right now. For the organizations that run such software, that don’t do it on 24/7-basis, but that still need fast turn-around time, buying grid-based horsepower in on-demand fashion will obviously be more cost efficient than keeping enough compute power laying around for when duty calls. So, there’s a quick market for some utility compute provider like Sun to clean up in. How much of this market was already buying Sun gear to address their HPC needs? And what will be Sun’s net gain if it corners the HPC market with its utility offerings? (In other words, how much HPC business will Sun take away from other competitors?) You know there’s overlap.
But longer term, if you follow Schwartz’s analogies, the destiny that awaits compute grids isn’t at all unlike the destiny that once awaited electricity, which, in another life, was only available to the wealthy few who could afford to run their own electric plant. Once the idea of a shared electric utility took hold, the cost of electricity was driven far enough down to the point that it made no sense for those wealthy few to run their own utilities.
In the aforementioned context, whether it’s a Monte Carlo simulation today, or running a Bloglines aggregation utility or all of Google tomorrow, the future compute grids shouldn’t discriminate against application type. To draw the comparison to electricity, their unique selling proposition must appeal to anyone who wants to scale in on-demand fashion. Imagine if, today, most of your appliances could draw power from the plug in the wall, but one — your VCR — needed its own generator. Yuck.
This is one reason that I think it will be interesting to watch how well Schwartz’s grid vision materializes. Early on, I agree that Sun will need to get some big wins on the HPC front, since that’s the type of computing that is currently best suited to the type of grid technology Sun has. But longer term, for Schwartz to scale Sun and to disrupt the computing business the way the electric grid forever changed the world, Sun and other utility vendors will have to make their grids available to a much larger class of applications. In other words, there’s a ceiling to where the HPC market will take any utility provider. The real growth opporutnity is in non-HPC applications. (Although I could probably make an argument for why Google and Bloglines will one-day be HPC applications.)
That grids-for-the-masses future will require one of two things: adapting grids to the non-HPC applications or adapting those applications to the HPC grid — neither of which are trivial. And even after those goals are reached, the Mark Fletchers of the world will still need to be convinced that it makes no sense to run their own plants. But, looking back at what Fletcher said, not only is he unconvinced, it’s clear to me that he’s barely paying attention to how a hardware utility might potentially be better way to run his business. In our discussion, I even tried to convince him that it might be. He isn’t alone. Marc Benioff, CEO of salesforce.com (another software utility), isn’t paying much attention either. Winning over these utility mavens, as I posited in my column, may be prerequisite to corporate culture adjusting itself in the way that Schwartz suggested it needs change in his most recent blog (a necessary change that I’m not disputing, by the way).
As I encounter more utility providers, I will continue my informal survey and report on the results.
