Breaking News
News & Analysis

Video: M'soft exec on federal research, multicore

12/3/2008 07:00 PM EST
1 Comment
NO RATINGS
More Related Links
View Comments: Oldest First | Newest First | Threaded View
Mapou
User Rank
Rookie
re: Video: M'soft exec on federal research, multicore
Mapou   12/4/2008 8:53:38 PM
NO RATINGS
Is Dan Reed for real? Any talk of hybrid or heterogeneous processors is a sign that Reed/Microsoft does not fully appreciate the radical nature of the multi/manycore paradigm shift that is around the corner. Programming hybrid processors is worse than pulling teeth with a crowbar. Why take that route? It does not make sense. On the contrary, logic dictates that we should lean toward a homogeneous/universal computing model. The stark reality is that the current computing paradigm is broken and it has been broken ever since Charles Babbage dreamt up his analytical engine. The industry still has the same basic idea of what a computer should be that Babbage had in his day. Parallel computing was the furthest thing from his mind. There can be no transition. What is needed is a complete break from the past and a dogged willingness to allow the failing legacy technology of the 20th century to wither into oblivion and museums. Otherwise we'll be spending huge sums of money to train a bunch of people to write expensive software that will not run on the parallel machines that are coming down the pike. Likewise, all this talk about F# and functional programming being well suited to multicore is a red herring. FP is counterintuitive and non-deterministic. It is a continuation of the same process/thread mentality that has gotten the industry into this mess in the first place. FP does not make easier to train millions of programmers for parallel programming jobs, I'm sorry. And, as I wrote in a response to Pat Gelsinger's video talk on EETimes yesterday, we can all forget about multithreading as a viable solution. You know, I am really coming to the point where I seriously doubt that either Intel or Microsoft can provide the type of leadership that is needed to get the industry around this difficult period in computer history. I am very disappointed. It is time for the baby boomer generation to step aside and let a new generation have a turn at the helm. You guys got us into this mess. Now you have run out of ideas simply because you are too old and set in your ways. If you're truly interested in solving this problem, as opposed to holding on to your job until you retire, Google "How to Solve the Parallel Programming Crisis" for more info.

Top Comments of the Week
August Cartoon Caption Winner!
August Cartoon Caption Winner!
"All the King's horses and all the KIng's men gave up on Humpty, so they handed the problem off to Engineering."
5 comments
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Radio
LATEST ARCHIVED BROADCAST
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.
Flash Poll