Language barrier
To take advantage of multicore chips, programmers will need software development systems that let them express themselves in fundamentally new ways.
To take advantage of multicore chips, programmers will need software development systems that let them express themselves in fundamentally new ways.
One key to making parallel algorithms efficient is to minimize the amount of communication between cores.
Charles Leiserson and his team are experts at designing parallel algorithms — including one for a chess-playing program that outperformed IBM’s Deep Blue.
Operating systems for multicore chips will need more information about their own performance — and more resources for addressing whatever problems arise.
Improving communication between distributed processors and managing shared data are two of the central challenges in creating tomorrow’s chips.