"I have a mind like a steel... uh... thingy." Patrick Logan's weblog.

Search This Blog

Tuesday, May 29, 2007

Pot-Kettle-Black

From cnet on the need for more concurrent programming...

Intel's Borkar said that Microsoft and other large software makers have known this shift is coming and have not moved fast enough.

"They talk; they talk a lot, but they are not doing much about it," he said in an interview following his discussion. "It's a big company (Microsoft) and so there is inertia."

Intel's main problem... they should have started becoming a software company a long time ago. In some ways, they have, but just not good enough.

Look at Sun. Sure Intel in significant ways has been and will be in a better position that Sun. But in other significant ways, I have to wonder: is Sun a software company or a hardware company? What about Intel? Which is better, for the long run?

There is no clear answer to me, except that Intel would be in a far better position if they were to take on more of Sun's strengths.

Maybe I am jaded as a former software developer for Intel. But there is so much more Intel could be doing to control their own destiny.

Borkar writes...

"Software has to double the amount of parallelism that it can support every two years."
Don't fall for this. I hope this was a meaningless aside attempt to make a fluffy analogy.

There is no Moore-ish path for software. Hardware essentially took the same von Neumann design and compressed it along Moore's predicted path, including increasingly more concurrency over time. The reduction in size and increase in performance was essentially due to better technology being applied to the same design up until the internet, heat, etc. forced more significant design changes.

And even now the design changes are essentially taking the same von Neumann design and replicating it on a chip.

Software? This is a different beast that will require more radical rethinking. Taking the internet and shrinking it down to everything, more or less.

2 comments:

fumanchu said...

Kinda reminds me of the "people can't think faster" discussion from Peopleware...if increased concurrency is needed, it needs to get a heck of a simpler interface first.

Patrick Logan said...

I don't see a lot of attention being paid to simplicity, "transactional memory" notwithstanding. :-)

Blog Archive

About Me

Portland, Oregon, United States
I'm usually writing from my favorite location on the planet, the pacific northwest of the u.s. I write for myself only and unless otherwise specified my posts here should not be taken as representing an official position of my employer. Contact me at my gee mail account, username patrickdlogan.