acm-header
Sign In

Communications of the ACM

ACM TechNews

Twins Separated at Birth: Cloud Computing and Hpc


View as: Print Mobile App Share:

Microsoft's Dan Reed is the director of the Cloud Computing Futures (CCF) organization, whose overall goal is the transformation of the way people construct and manage very large-scale computational resources. "The CCF story is about approaching cloud computing infrastructure as an integrated design problem, looking at the balance of support infrastructure, computing hardware and software, not only at a single site but across an international network of interconnected sites," Reed says. He sees similarities between cloud computing and high-performance computing (HPC) beyond scale, and observes that "the specific aspect relative to HPC is that cloud services are game changers, just as commodity clusters were a decade ago and graphics accelerators have been recently."

Reed describes the design and construction of mega-scale data centers and current petascale and future exascale systems as "twins separated at birth," and lists interconnects, memory and storage hierarchies, and the future of non-volatile memories, heterogeneous multicore processors, system reliability and resilience, packaging, cooling and energy efficiency, and programming as shared design challenges. He says that CCF's objective is to inform the development of the future cloud infrastructure by drawing on new and emerging hardware and software technologies. Reed says clouds should not replace HPC, at least at the very high end, but should serve as a tool to more effectively underpin the rest of the infrastructure. He says clouds can act as a catalyst for new kinds of research, and concludes that cloud computing "can provide seamless extension of familiar desktop tools and interfaces, allowing computing and analysis to scale within the same environment that researchers use every day."

From HPC Wire
View Full Article


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account