Supercomputers are everywhere, but they hardly get the attention that they deserve. Hundreds of these huge machines churn out all sorts of data and analysis that many humans use in their everyday life, but most take it for granted. By and large, supercomputers have been "stale" and "boring" to average consumers. After all, you can't exactly buy and use your own supercomputer (at least not without tons of cash and a really good application), but Virginia Tech's new supercomputer
As the worlds of the CPU and GPU
have collided over the years, the same thing now seems to be happening in the supercomputer realm. The university has acquired a $2 million grant that will allow them to build the HokieSpeed, but what makes this unit stand out is the combination of GPUs and CPUs. Usually, supercomputers rely almost entirely on CPU
power. But this one is a "new heterogeneous supercomputing instrument based on a combination of central processing units (CPUs) and GPUs." In other words, this is one of the first supercomputers where the GPU is just as important as the CPU.
The public is being told that this new machine "is expected to deliver 35 times better peak performance, 70 times better peak power efficiency, and 14 times better peak space efficiency than System X." Wu Feng, associate professor of computer science as well as electrical and computer engineering and principal investigator on the grant, described the HokieSpeed as such: "HokieSpeed is expected to catalyze new approaches for conducting research via the synergistic amalgamation of heterogeneous supercomputing and cyber-enabled tools to enhance ease of use. In particular, it will give end users the ability to perform in-situ visualization for rapid visual information synthesis and analysis. It will also control their level of immersion in the discovery process - from being completely immersed, making real-time intuitive decisions via a large-scale gigapixel display, to observing the instrument automatically collect, organize, and analyze data in support of visual analytics. A large set of Virginia Tech researchers from across the university will be actively involved in research using HokieSpeed."