Joe Landman from Scalability.org writes that customers expecting a magic bullet from HPC in the Cloud will face disappointment:
For the economics of HPC in the cloud to work, you need to be at or below the price performance knee, where you can minimize the system cost while maximizing the cycles used. You may be able to charge more for faster cycles, but in general, most people don’t want to pay for much faster.
Joe goes on to say that the needs of IT managers (who need to amortize their costs over three years) and Cloud users (who want to have the cheapest, most efficient cycles from the latest processors and largest memories) are fundamentally at odds.
Comments
Trackbacks