Ever since the clock speeds of individual computer chips stopped doubling, those who have wanted to run existing applications faster -- such as algorithmic trading desks -- have had to turn to grid computing or specialized hardware to get that extra processing power. Multicore chips (which put two, four or more computer processors on one chip) and clusters of computers hold out the promise of high-performance computing but can't truly provide it to the typical application. The reason for this, as we've noted before, is that most applications have been written in a single-threaded manner -- they're designed to do one thing at a time. For an application to take advantage of more than one processor, whether on a multicore chip or across clusters of computers, it needs to be rewritten so that it's multi-threaded (in other words, it can perform several tasks at the same time) or optimized in some other way.Enter SIMtone, whose Aspeed software enables single-threaded software to act like it's multi-threaded by providing an API wrapper that enables functions to run as separate processes on multiple machines. The original application runs as normal, unaware that certain tasks are taking place on different computers, but faster because it's multitasking. SIMtone yesterday announced support for Microsoft's latest cluster server, Windows HPC Server 2008. So now, those that want to run existing applications across a cluster of Windows servers and see some speed benefits, have a way to do so.