Resumen
On December 16, 1996, a sustained rate of 1 teraflops was achieved by "ASCI Red," a system employing some 7,000 Intel Pentium Pro processors at Sandia National Laboratory in New Mexico. Much work remains to achieve Tflops per second rates on a broad range of scientific applications. Following the custom of marking advances in computing by factors of 1,000, the next major milestone is a sustained rate of 1 petaflops. In addition to prodigiously high computational performance, such systems must, out of necessity, feature very large main memories, between 10 Tbytes and 1 Pbyte depending on application, as well as commensurate input/output bandwidth and huge mass storage facilities. The current consensus of scientists who have performed initial studies in this field is that "affordable" petaflops systems which may be feasible by the year 2010, assuming that certain key technologies continue to progress at current rates. Demand for state-of-the-art computing power appears insatiable. There are a number of difficult technical problems that need to be solved in the next few years to achieve the goal of petaflops computers by the year 2010. INSET: Petaflops Research Questions.