Complex Insight - Understanding our world

The advent of Intel's massively parallel coprocessor will make every server a supercomputer. This week, Intel unveiled its new Xeon Phi coprocessor, which puts an astonishing 50 x86 cores onto a single PCI-connected card. The term "coprocessor" should be understood in context. Every one of the Phi's cores can boot Linux and run any x86 software. However, the card itself needs to plug into a system that has an independent CPU, which basically oversees the Phi's operations. Hence, the coprocessor appellation. The first model to be released in Q1 of next year will have 50 cores, and the follow-up coprocessor slated for release in mid-2013 will have 60 cores. Each processor supports four threads, making for 200 threads for the initial Phi. The cores run at 1.05 GHz and sport a 512-KB L2 cache each. They collectively share 8 GB of GDDR5 memory. They will liely compete with GPU based solutions from NVIDIA and AMD - but with a more familiar programmign model. Things just got very interesting in the high performance compute world. Click on the image or the title to learn more.

Wired has an interesting article on mobile computing. As ARM seeks to put cell phone chips into our supercomputers, Intel is doing the reverse. The lines between the mobile hardware and data center hardware are blurring. Intel’s experimental Single-chip Cloud Computer project, or SCC is a 48 core chip acts as a “network” of processors on a single chip, with two cores per node. The nodes actually communicate to each other much the same way nodes in a cluster in a data center would.  While these types of architectures are today used in big data applications that run on a server cloud environment there are actually many cases where a hybrid model would make the most sense including artificial intelligence, machine vision and augmented reality applications.  Click on the image or the title to learn more.

Researchers from Johns Hopkins and Northwestern universities have discovered how to control the shape of nanoparticles that move DNA through the body and have shown that the shapes of these carriers may make a big difference in how well they work in treating cancer and other diseases.  The use of computer models allowed Luijten’s team to mimic traditional lab experiments at a far faster pace. These molecular dynamic simulations were performed on Quest, Northwestern’s high-performance computing system. The computations were so complex that some of them required 96 computer processors working simultaneously for one month.  “Our computer simulations and theoretical model have provided a mechanistic understanding, identifying what is responsible for this shape change,” Associate Professor Eric Luijten said. “We now can predict precisely how to choose the nanoparticle components if one wants to obtain a certain shape.”. Click on the image or the title to learn more.

Purdue University is working with semiconductor researchers, including Intel research scientist Charles Augustine of its Circuits Research Lab (Hillsboro, Ore), to develop spin-based neuromorphic microchips as the ultimate parallel...

the world’s third-fastest supercomputer --known as Mira--will complete tests of its new upgraded software and begin running the largest cosmological simulations ever performed at Argonne National Laboratory. One point of interest is the HACC (for Hardware/Hybrid Accelerated Cosmology Code) the underpinning code base is modular and reconfigurable for different compute architectures including hybrid CPU and GPU architectures. Links for further information in the article sidebar. Click on the image or the title to learn more.

Titan is the latest supercomputer to be deployed at Oak Ridge National Laboratory (ORNL), although it's technically a significant upgrade rather than a brand new installation. Jaguar, the supercomputer being upgraded, featured 18,688 compute nodes - each with a 12-core AMD Opteron CPU. Titan takes the Jaguar base, maintaining the same number of compute nodes, but moves to 16-core Opteron CPUs paired with an NVIDIA Kepler K20 GPU per node. The result is 18,688 CPUs and 18,688 GPUs, all networked together to make a supercomputer that should be capable of landing at or near the top of the TOP500 list. Click on the image or the title to learn more