Matt\’s Cuppa

My take on tea, technology, and our environment


Posted by telecommatt on March 9, 2007

Greg Matter : Weblog


And, no, I’m not paraphrasing something that I bet Thomas J. Watson never uttered in 1943 anyway. But he should have because, ultimately, he might turn out to have been right.

Although his view is slightly tainted by the fact that he is with Sun, I happen to think the guy is right. It may or may not be five computers in the end, but centralized computing is fast becoming nonsensical. The sheer amount of data that is processed on a daily basis for even a mid-level company has grown exponentially in the past decade six minutes. His point is that computing is going to look a lot like the energy sector. A few big giants and some niche market players. While I agree that the players will end up consolidating, I think the actual “computer” will be designed more like a large scale telephony transport network in reverse. Your data network is designed to distribute information from one source to another (or many others). A distributed processing network is going to be harnessing processing power from various sources and consolidating that processing power to complete a task. See, opposite directions, same structure. I wonder who will be the Big Five. Can you imagine the kind of processing power you’d see from The Chinese Computer?


  1. Just some follow up thoughts from last night’s post. My brain doesn’t always catch everything at the first pass when I’m writing at midnight.

    The whole application of distributed computing is certainly nothing new, and has been used for years in, for example, the academic world. The SETI program is perhaps the most iconic in this setting. How does that apply to the consolidation of computing companies refered to in the Matter article? The answer is simple. As companies consolidate and aquire disparate resources, these companies will need to be able to utilize the processing power of the company they have acquired. This isn’t about tossing it all in a box and shipping it somewhere. It’s about leaving everything where it is because it’s already WIRED. There is already a communications network in place. The node/leg setup of a typical transport network is a perfect structure for a distributed computing network to piggyback on. Everything is there, duplex communication, largest to smallest architecture, redundancy, etc. The question goes down the network and the answer goes up, carried by the existing network protocol. What’s my point? Aside from market considerations and economies of scale, the other factor that makes the scenario in the Matter article a likely future is that the infrastructure is already in place.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: