A recent online survey by Forrester Research shows that only a quarter of IT execs at 149 large North American companies are using grid computing for its intended purpose: the processing of huge amounts of scientific or engineering data. The rest, well, they're using their idea of grid computing for just about everything else.
For example, 57 percent are using it to share data across computers and 53 percent say they're using it with databases and other back-office software.
Grid computing in recent years has often been described as distributed computing, which has blurred its true definition as the sharing of CPU resources across a network, so that all machines function as one large supercomputer.
As a result, a fifth of the respondents to the Forrester survey said grid computing has different meanings. About a third said it refers to clusters of computers, sharing data across machines or massively parallel processing.