The case for more efficient computing is clear, as Dr. Alistair Dunlop presents in his article "Is your grid burning money":

"A common misconception is that money spent optimizing could be better spent on newer, faster or simply more servers. In a grid nvironment, the applications are spread across many servers, with the result that relatively small optimizations could have a large overall impact. Consider in the first instance application optimizations. Obtaining even a modest 20 percent improvement in execution time through choice of processor architectures or high-performance libraries would equate to a saving of more than $300,000 per annum on a 500 node single line-of-business grid.

Optimizations that affect the entire grid, such as grid aggregation or modification to the scheduling algorithm, can result in efficiency improvements of 20 percent or more. Taken over a 10,000 CPU grid, this small change could save in the order of $6 million each year. Combined with application optimizations, savings of around $10 million annually are possible. Cost savings are only half the story. The real value is often taken to be the opportunities that can be realized as a result of the efficiency improvements."

The improvements of 20% can be achieved traditional means and depending on the application domain an accelerated systems can bring performance increases of up-to 200x.

Any business or industry can benefit from improvements to their infrastructure, either in terms of cost effectiveness or absolute performance but some are more dependent on computational power so that the business case for the switch to use heterogeneous computing is more compelling. These include: