The Ramp Project’s Goal Is to Enable the Intensive, Multidisciplinary Innovation That the Computing Industry Will Need to Tackle the Problems of Parallel

......In 2005, the computer hardware industry took a historic change of direction: The major microprocessor companies all announced that their future products would be single-chip multiprocessors, and that future performance improvements would rely on software-specified parallelism rather than additional software-transparent parallelism extracted automatically by the microarchitecture. Several of us discussed this milestone at the International Symposium on Computer Architecture (ISCA) in June 2005. We were struck that a multibilliondollar industry would bet its future on solving the general-purpose parallel computing problem, when so many have previously attempted but failed to provide a satisfactory approach. To tackle the parallel processing problem, our industry urgently needs innovative solutions, which in turn require extensive codevelopment of hardware and software. However, this type of innovation currently gets bogged down in the traditional development cycle: N Prototyping a new architecture in hardware takes approximately four years and many millions of dollars, even at only research quality. N Software engineers are ineffective until the new hardware actually shows up, because simulators are too slow to support serious software development activities. Software engineers tend to innovate only after hardware arrives. N Feedback from software engineers on the current production hardware cannot help the immediate next generation because of overlapped hardware development cycles. Instead, the feedback loop can take several hardware generations to close fully.