Michael Wolfe, Compiler Engineer at The Portland Group, Inc., wrote an article last week on many available options that claim to make parallel programming easy, discusses complexity of parallelism, and offers a few methods for solving the current “parallelism crisis”.
This week, he writes about GPU architecture and applications for GPUs.
One of the most exciting developments in parallel programming over the past few years has been the availability and advancement of programmable graphics cards. A high end graphics card costs less than a high end CPU and provides tantalizing peak performance approaching, or exceeding, one teraflop. Since microprocessor peak performance tops out at about 25 gigaflops/core (single precision), this potential, at such low cost, is worth exploring. Harnessing this performance, however, is problematic.
An interesting read.
