I checked this article on EE times and I was like why the hell hasn’t I thought of this before? Well actually, I did think of it before and when I asked my professor about it … he said its probably way too expensive and inefficient to make.
So what is this about? The answer is too technical but just to make it simple. You know how Intel Core Duo is better than an Intel Core 1. If you have two processors, it will be faster than one. Thats called parallel processing. You can do the same with memories. Memories are one of the biggest bottlenecks in high performance computing and the ability to access memory cells in parallel will mean an improvement in accessing times, and therefore improved overall system performance.