So I was thinking about paralllel processing… You know.. Just the kind of thing that I do while riding. Was wondering where and how the processing power will be increased in the future so that the semiconductor industry can adhere to “Moore’s Law“. Seemingly, over time, it is going to get more and more difficult for them to reduce the size of the processor. So i’m guessing they would start increasing the size. Confused?
Well, let me explain. I believe we have already seen the first generation of parallel processors. The current quad core, dual core are the first examples of it. But, lets say tomorrow, we get a 16 core processor ( I am assessing its already there? ). So now, are programs making use of this? Since most of the computer programs that are currently running are built for single processor systems, ofcourse they wont be making use of this feature.
But since we already know that tomorrows processors are going to be multi-processors, shouldn’t the programming languages anticipate that and provide constructs so that if ever such proccessors come up in the future, the programs should be able to make use of them?
We humans, have a dream. Atleast we programmers , I assume, do . Why cant we make the current computer as “smart” as the brain. So can we achieve the levels of a human brains ( or atleast a rats? ) with a huge multiprocessing system?
But then, writing a system that can work in a multiprocessing environment is not a trivial task. The mental acumen required to do that is not trivial either. But the traits of humans is that we are not born geniuses. We learn. So if even one person in the world deciphers an easy way to do it, the rest of us will follow through. But then again, who will bell the cat , huh!!