Nvidia VP Bill Dally has claimed that "Moore's law is dead" and declares that ground-up parallel processing is the future of computing.

Moore's law describes that trend for computing performance – processing speed, memory capacity and the like - to double approximately every two years, named after Intel co-founder Gordon E. Moore, who first outlined his theory back in 1965.

Moore's law is over?

However, writing a guest column for Forbes magazine this month, Nvidia man VP Bill Dally thinks that "Moore's law is dead."

Dally says that dual, quad- and hex-core solutions are becoming increasingly inefficient and likens multi-core chips to "trying to build an airplane by putting wings on a train."

Instead, Nvidia feels that its own approach – ground-up parallel solutions made for increased energy efficiency – is the answer.

Energy efficiency is key

"Going forward, the critical need is to build energy-efficient parallel computers, sometimes called throughput computers, in which many processing cores, each optimized for efficiency, not serial speed, work together on the solution of a problem," Dally writes.

"A fundamental advantage of parallel computers is that they efficiently turn more transistors into more performance. Doubling the number of processors causes many programs to go twice as fast. In contrast, doubling the number of transistors in a serial CPU results in a very modest increase in performance–at a tremendous expense in energy."

Nvidia's own CUDA architecture is already found in GeForce, ION, Quadro and Tesla GPUs.

"The path toward parallel computing will not be easy. After 40 years of serial programming, there is enormous resistance to change, since it requires a break with longstanding practices. Converting the enormous volume of existing serial programs to run in parallel is a formidable task, and one that is made even more difficult by the scarcity of programmers trained in parallel programming," the Nvidia man continues.

"The good news is that there is a way out of this crisis. Parallel computing can resurrect Moore's Law and provide a platform for future economic growth and commercial innovation. The challenge is for the computing industry to drop practices that have been in use for decades and adapt to this new platform."

Read more: http://www.techradar.com/news/computing ... z0mrkvWkIw

Parallel Processing in Computers

Main article: Parallel computing

The simultaneous use of more than one CPU or processor core to execute a program or multiple computational threads. Ideally, parallel processing makes programs run faster because there are more engines (CPUs or cores) running it. In practice, it is often difficult to divide a program in such a way that separate CPUs or cores can execute different portions without interfering with each other. Most computers have just one CPU, but some models have several, and multi-core processor chips are becoming the norm. There are even computers with thousands of CPUs.

With single-CPU, single-core computers, it is possible to perform parallel processing by connecting the computers in a network. However, this type of parallel processing requires very sophisticated software called distributed processing software.

Note that parallel processing differs from multitasking, in which a CPU provides the illusion of simultaneously executing instructions from multiple different programs by rapidly switching between them, or "interleaving" their instructions.

Parallel processing is also called parallel computing. In the quest of cheaper computing alternatives parallel processing provides a viable option. The idle time of processor cycles across network can be used effectively by sophisticated distributed computing software.

Parallel computing is very hard to program in. I had a friend who was a contract programmer in a number of languages and he used to program in Occam and he said that it was the worst language to program in because you have to keep track of too many processes.

I'd guess the human factor is deadly in any sort of programming

It's the kind of thing I could see Apple doing one nay. Creating a parallelised version of the OS and forcing all of their developers to port their apps to the new architecture. Apple would probably create some method to run existing applications with reduced performance and capabilities...

People I know who have done multi-threaded programming hate it because it is so challenging on the mind.

I would imagine that any such future OS that is fully parallel will be such that the app programmers do not have to learn multi processor languages. I expect that there will be compilers that convert the program into a form that the OS can use.

People I know who have done multi-threaded programming hate it because it is so challenging on the mind.

the number of transistors that can be placed inexpensively on an integrated circuit has doubled approximately every two years