<img src=”http://media2.hpcwire.com/hpcwire/Cloud_Storage_and_Bioinformatics_in_a_private_cloud_Fig._3_150x.png” alt=”” width=”95″ height=”95″ />The top research stories of the week include an evaluation of sparse matrix multiplication performance on Xeon Phi versus four other architectures; a survey of HPC energy efficiency; performance modeling of OpenMP, MPI and hybrid scientific applications using weak scaling; an exploration of anywhere, anytime cluster monitoring; and a framework for data-intensive cloud storage.
<img style=”float: left;” src=”http://media2.hpcwire.com/hpcwire/Steven_Chu_official_DOE_portrait_150x.jpg” alt=”” width=”94″ height=”92″ />US Energy Secretary Steven Chu steps down after a controversial term in which he championed high performance computing, launched dozens of energy research centers, and led the government’s attempts to help industry transform the country’s energy landscape. But his most famous decision was the most politically divisive: backing a company called Solyndra.
A new study that contradicts the influential EPA report about a doubling of data center energy consumption made waves this week.
According to HP, we are transitioning from the Information Age to the Insight Age, which signals a need for new architectures.
Over the next few years, the very sources of massive energy consumption–data centers that power cloud computing–could soon turn to serve the goals of energy efficiency programs. Cloud platforms and the scalable needs they can serve are increasingly being seen as key to energy management, especially as the range of devices, sensors and measurement tools grows and renewable sources are being fed back into the system.
Exascale computing promises incredible science breakthroughs, but it won’t come easily, and it won’t come free.
Recent research from the University of Melbourne indicates that in many cases, the cloud is not the “greenest” option due to data transfer energy consumption. The authors argue that prior studies have over looked transfer in their findings.
Power consumption of big supercomputers not that outrageous.
Exascale computing. What is it good for? Certainly not to solve problems that need solving today.
It seems like since 1993, when the Internet got such a foothold on the imagination of the world, that there’s always been the rumor of the next great disruptive technology lurking right around the corner.