3D scaling could be important if there is high interdependence between the operations of each CPU, but I maintain that we may have to build new computing models to really take advantage of systems built around CPUs like this. Separate memory for each processor is a good start, since it relieves the bottleneck of access to shared memory. Data centers are a good application for this, because they often work on large numbers of independent transactions. By the way, we may also have to look at the nomenclature. I just realized that this isn't really a Central Processing Unit (CPU). It's more like a Central Processing Pool (CPP?). For that matter, is the 'processor' the chip or each of the sections of it? The same word is used for both in this article.
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.