“It was indicated in my keynote this morning there are two really fundamental challenges we’re facing in the next two years in all sorts of computing – from supercomputers to cell phones. The first is that of energy efficiency. With the end of Dennard scaling, we’re no longer getting a big improvement in performance per watt from each technology generation. The performance improvement has dropped from a factor of 2.8 x back when we used to scale supply voltage with each new generation, now to about 1.3 x in the post-Dennard era. With this comes a real challenge for us to come up with architecture techniques and circuit techniques for better performance per watt.”
In this video from the 2015 Hot Chips Conference, Charles Zhang from Phytium presents: Mars – A 64-Core ARMv8 Processor. Formed in China in 2012, Phytium is a unique technology provider of HPC servers, focusing mainly on high performance general microprocessor, accelerator chip, reference board design and various servers design from blade, cluster, standard stack to HPC Server. “Optimized for HPC, the Mars chip features eight panels, each with eight “Xiaomi” cores. The panels share an L2 cache of 32 MB, two Directory Control Units and a routing cell for the internal mesh.”
Today the European Consortium announced a step towards Exascale computing with the ExaNeSt project. Funded by the Horizon 2020 initiative, ExaNeSt plans to build its first straw man prototype in 2016. The Consortium consists of twelve partners, each of which has expertise in a core technology needed for innovation to reach Exascale. ExaNeSt takes the sensible, integrated approach of co-designing the hardware and software, enabling the prototype to run real-life evaluations, facilitating its scalability and maturity into this decade and beyond.
Today European datacenter specialist DATA4 Group and Qarnot Computing announced a new type of distributed computing system that offers “greener and more efficient computing services.” The system is based on Qarnot’s Q.rad, a smart and connected digital heater. “Think of the device as the fusion of an electrical heater and a server. In the Q.rad model of computing, servers are placed in rooms that need heat. They are then networked together to form a physically distributed cloud computing resource.”
Today Centerprise International (Ci) in the UK announced a collaboration with E4 Computer Engineering to develop next-generation datacenter technologies for HPC. “This is an exciting development for both companies, as it combines the specialist knowledge of E4 in the field of high performance computing with our considerable experience in building quality, customized hardware solutions and our expansive reach in the UK IT channel,” said Jeremy Nash, Centerprise Sales Director.”
Although liquid cooling is considered by many to be the future for data centers, the fact remains that there are some who do not yet need to make a full transformation to liquid cooling. Others are restricted until the next budget cycle. Whatever the reason, new technologies like Internal Loop are more affordable than liquid cooling and can replaces less efficient air coolers. This enables HPC data centers to still utilize the highest performing CPUs and GPUs.
A new paper from ORNL’s Sparsh Mittal presents a survey of approximate computing techniques. Recently published in ACM Computing Surveys 2016, A Survey Of Techniques for Approximate Computing reviews nearly 85 papers on this increasingly hot topic.
Penguin Computing has renewed as a Platinum Member of Open Compute Project (OCP). Leading with the OCP-based Tundra Extreme Scale (ES) Series, Penguin was recently awarded the CTS-1 contract with the NNSA to bolster computing for national security at Los Alamos, Sandia and Lawrence Livermore national laboratories.
The EMiT 2016 Emerging Technologies Conference has issued its Call for Papers. Hosted by the Mont-Blanc project and the Barcelona Supercomputing Centre, the event takes place June 2-3, 2016 in Barcelona.
Buddy Bland from ORNL presented this talk at SC15. “Summit will deliver more than five times the computational performance of Titan’s 18,688 nodes, using only approximately 3,400 nodes. Each Summit node will contain multiple IBM POWER9 CPUs and NVIDIA Volta GPUs all connected together with NVIDIA’s high-speed NVLink and a huge amount of memory.”