Google, the world's most popular search engine and a leading online service provider, has acquired PeakStream, a startup founded back in 2005 that specialized on development of tools that relied on computing power of graphics processing units (GPUs) as well as multi-core central processing units (CPUs).

?We believe the PeakStream team's broad technical expertise can help build products and features that will benefit our users. We look forward to providing them with additional resources as they continue developing high-performance applications for modern multi-core systems,?a statement by Google reads.

The only product that PeakStream developed is PeakStream platform, a set of tools that allows to rapidly create software applications that rely on GPUs or multi-core CPUs. The officially announced PeakStream version supported ATI Radeon X1950 graphics cards only.

The move demonstrates that after many years of talking, a serious market player have paid attention to the so-called general purpose computing on GPUs (GPGPU) and even though PeakStream also worked on tools that maximized performance of various applications on multi-core CPUs, their primarily goal was to use GPUs for general computing.

Google is unlikely to continue selling PeakStream products to third parties, but would rather use the tools when making its own software, which will run faster compared to programs developed without GPGPU or multi-core microprocessors in mind.

According to estimations of ATI, graphics product group of Advanced Micro Devices, using GPU processing of risk assessment models similar to those used by financial institutions?are completed 16 times faster than traditional methods (based on simulations performed by PeakStream using traditional processing models vs. using ATI processors.), oil and gas companies may see seismic model processing increased by more than 20 times (based on results by PeakStream), and Stanford University is seeing disease research accelerated by as much as 40 times, giving them the ability to process three years worth of research data in just one month (Based on Stanford University's testing that compares work unit processing on an Intel Pentium 4 2.80GHz processor vs. similar processing being conducted on a Radeon X1900 XT).

Financial terms of the transaction were not disclosed.