The emergence and migration of new workloads and usage models to mainstream computing will put enormous demands on future computing platforms, including demands for: | | Much higher performance | | | Much lower power density | | | Greatly expanded functionality | |
Given these seismic shifts in computing usage, how we define and architect future computing platforms has to change dramatically. These changes must holistically comprehend and satisfy not only computation, but also interface and infrastructure requirements. Intel‘s long-range vision for the collective evolution of these three fundamental platform elements, and the architectural innovation and core competencies enabling that evolution, is what we call Platform 2015. Looking forward, the microprocessor, as the key determiner of computational capabilities, will only increase in importance in driving and defining the platforms based on it. The reason? The microprocessor itself is on the verge of undergoing significant transformation, requiring new advances in addition to achieving the predictions of Moore‘s Law. Continuing increases in transistor count are vital, of course, but the next leap requires us to rethink the basic foundations, such as process technology, architecture and software. |