Scientists at IBM speculate that the world’s most powerful supercomputer processors could shrink to the size of a sugar cube. They see many computer processors stacked on top of one another, cooling them with water flowing between each one. The plan is to reduce computers’ energy use, rather than just to shrink them. Most of the computer’s energy costs is in cooling, because computing power generates so much heat as a side product. Dr Bruno Michel and his colleagues have already built a prototype to demonstrate the water-cooling principle. Called Aquasar, it occupies a rack larger than a refrigerator.
In the past, computers were dominated by hardware costs. Future computer costs will hinge on green credentials rather than speed.”
In this special guest feature, Rob Farber writes that a study done by Kyoto University Graduate School of Medicine shows that code modernization can help Intel Xeon processors outperform GPUs on machine learning code. "The Kyoto results demonstrate that modern multicore processing technology now matches or exceeds GPU machine-learning performance, but equivalently optimized software is required to perform a fair benchmark comparison. For historical reasons, many software packages like Theano lacked optimized multicore code as all the open source effort had been put into optimizing the GPU code paths." [Read More...]