AMD is talking today about its single-system, Accelerated Computing platform “that breaks the teraflop computing barrier.”
At a press event in San Francisco, AMD demonstrated a “Teraflop in a Box” system running a standard version of Microsoft® Windows® XP Professional that harnessed the power of AMD Opteron™ dual-core processor technology and two next-generation AMD R600 Stream Processors capable of performing more than 1 trillion floating-point calculations per second using a general “multiply-add” (MADD) calculation. This achievement represents a ten-fold performance increase over today’s high-performance server platforms, which deliver approximately 100 billion calculations per second.
The press release has a note that irritates me (if you know that this isn’t wrong, leave a comment so I don’t continue to labor in ignorance). In answer to the question “what’s a teraflop” their release says: “A teraflop is one trillion floating point operations per second.” This is incorrect as there is no such thing as a teraflop. It’s teraflops, as in tera floating point operations per second.
FLOPS is both singular and plural, unless you seriously want to argue for FLOPSES, at which point I’ll have to strike you hard between the eyes.