Intel Tera – not firma, but coming Here is a long-term question: is Intel starting the long walk away from the x86 architecture and towards what it sees as the "next big thing"? There is certainly no immediate need to panic. Nothing is going to change for a good few years yet, but there has to come a time when the aged and venerable x86 fundamentals are put out …
There is already a parallel-processing language with well-developed compilers, etc. The language was occam, originally developed for the transputer. Inmos also produced a version of C which was in some ways easier to use than native occam. Occam and the transputer provided parallel execution threads and simultaneous/independent processing and i/o. Parallel processing is not new, and it is does not need a new programming paradigm, it's more a matter of thinking of old problems in a new processing topology.
Having worked with Inmos Transputers and Intel i860s in the 90s this does all sound rather familliar.
However, although there defintely is a need for a parallel programming laguage for high performance applications, for most needs this is a red herring.
Modern operating systems run many separate processes to perform individual tasks, even Windows. So once the kernel can handle a a large number of processors (not necessarily a trivial task), the general application layer can carry on as before with the many processed geninely getting run in parallel.
Indeed, when I was doing parallel simulations it was impossible to beat the overall efficiency of running a complete simulation on each processor, rather than running a succession of parallel jobs really quickly.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
