This article is <i>why</i> I read Hacker News!<p>To surpass limitations; and see limitations surpassed, in all areas of technology.<p>This article exemplifies that ethos.<p>This is the future, or rather a glimpse of it, today.<p>Well done!
So what is this memory wall?<p>> this concept revolves around the idea that computer processing units (CPUs) are advancing at a fast enough pace that will leave memory (RAM) stagnant<p>> According to Moore’s Law, which states that the number of transistors in a circuit doubles every two years, CPUs will eventually become too fast to yield any noticeable difference in computing speed. Once we reach this so-called memory wall, program/app execution time will depend almost entirely on the speed at which RAM can send data to the CPU. So even if you have an incredibly fast processor in your computer, it’s function may be limited to the speed of your RAM.<p>Oh, it's utter nonsense.<p>For one, it's ignoring that the cache speeds scale right along with the CPU. You could get rid of DRAM entirely to prove there is no 'wall'.<p>But what really kills the argument is a proper analysis of how memory is set up. Internal to the RAM chip, there's a ridiculous amount of bandwidth. The bottleneck is using a 64-bit-wide bus to communicate with the CPU. But that bottleneck is only a cost saving measure <i>on top</i> of Moore's Law. If we back off to pure Moore's Law, we can quite easily add more and more memory width/channels and not hit any limits.