Download now Free registration required
Companies such as Google and Microsoft routinely process many terabytes of data, and users of desktop search engines routinely pose queries across hundreds of gigabytes of data stored on their hard drives. There's no reason to expect that the appetite for collecting and processing data will stop growing at its current breakneck speed. To satiate the appetite for large-scale data processing, current technology must overcome three challenges. First, the recent slowdown in uniprocessor performance and the difficulties in programming their Chip Multiprocessor (CMP) replacements make it increasingly difficult to bring large computing power to bear on a single problem.
- Format: PDF
- Size: 750.47 KB