Sponsored Feature Read this Register interview to find out what Professor Onur Mutlu of ETH Zurich thinks about the computing architecture we should use to meet the demands of the next generation.
Cloud Adoption Basics: The Fundamentals of Cloud Architecture
The Registry: If we look at how computer architecture works today, what are the challenges and shortcomings of this model?
Professor Onur Mutlu: There are major problems with how we design computer systems today. One of the biggest is that even though datasets are growing, and we're trying to do more sophisticated things with them, the components that do the actual computation are a very small part of the system. If you take a typical node, more than 98 percent of that node is dedicated to storing, controlling, and moving data, while the processors working on that data are a very small part of it. The way we design our systems is very processor-centric. The processor is king, and everything must be moved to it so that you can perform calculations. The storage system, the memory system, and the interconnects are not active components working productively on that computation. When you are continuously moving data between the processor and the memory or storage subsystem, it represents a major bottleneck.
Reg: How does this fit with the age of data-heavy applications?
