Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


Archives for August 2011

Video: In-Booth Racking Contest at VMworld

Do you think you can rack up servers faster than the next cluster monkey? In this video, contestants race to build a rack from foam components in the Cisco booth at VMworld. Great fun! Personally, I think we did this one better at SC2003 in Phoenix, where SDSC built a live, 128-node Rocks cluster in […]

IBM Opens Power8 Kimono – Wafer Baked in 22 Nanometers

By Timothy Prickett Morgan • Get more from this author Last May El Reg gave you the little information we were able to gather on the impending Power7+ and future Power8 processor designs and their possible announcement dates in Power Systems machines. Data was a little thin, and intentionally so on the part of Big Blue. But […]

Scalable Informatics Jackrabbit Storage Delivers "Raw, Unapolgetic Firepower"

Scalable Informatics’ Joe Landman writes that the company’s new Jackrabbit is a 5U 108TB tightly coupled storage unit with amazing performance for spinning disk–sustained 4.1 GB/s writes and 5.2 GB/s reads for 1 TB sized files on a single machine with local disk: This is “spinning rust.” This is not SSD/Flash. I think this just […]

IBM to Bounce Back with Mira and Sequoia, Powered by Blue Gene/Q

Datacenter Knowledge takes a sneak peek at IBM Blue Gene/Q supers in the works: the 10 petaflop Mira system at Argonne National Lab and the 20 petaflop Sequoia at Lawrence Livermore. The Power-7 chip was set to perform at 256 gigaflops per 8 cores and consume 200 watts, where the Blue Gene/Q SoC will pull […]

EMC Breaks Record with 1 Million IOPS in VMware

This week EMC announced record-breaking storage throughput and bandwidth in a VMware vSphere 5 environment. The company described the achievement as accelerating customers’ journey to the cloud. With this new storage performance level, organizations can now confidently virtualize even their largest online transaction processing (OLTP) applications, with the most stringent workloads and lowest latency service […]

Video: Birth of Milky Way

In this video, a supercomputing simulation depicts the physics involved in the formation of the Milky Way. Researchers at the University of California, Santa Cruz and the Institute for Theoretical Physics in Zurich used over 8 months of supercomputing time to run the simulation, called “Eris,” which solves a long-standing problem that led some to […]

New Intel Fed Subsidiary to Focus on Exascale

Today Intel announced the formation of Intel Federal LLC, a wholly-owned subsidiary focused on HPC and Exascale computing. Reaching supercomputer performance levels of a hundred times more powerful than today by 2018 will require the combined efforts of both industry and government. An ExaFlop supercomputer’s performance is the equivalent of every person on Earth making […]

SGI Looks to Expand in China

China Daily is reporting that SGI plans to double the size of its Chinese workforce in anticipation of a fivefold increase in business in the region over the next three years. As the fourth-largest high-performance computing and cloud-computing equipment provider worldwide by sales revenue, SGI said it will add a new office in the southwest […]

RCE Podcast: AMD Core Math Library ACML

In this RCE Podcast, Brock Palen and Jeff Squyres speak with Chip Freitag of AMD about the AMD Core Math Library an high performance implementation of the BLAS and LAPACK in addition to FFT’s and PRNG’s. Download the MP3.

Video: Hurricane Irene Storm Tide Simulation for Chesapeake Bay

In this video generated by researchers at the Virginia Institute of Marine Science, a supercomputing simulation shows a 30-hour forecast of storm-tide water levels during the passage of Hurricane Irene past the mouth of Chesapeake Bay. The simulation begins on Saturday, August 27th, 2011 at 12:00 hours Greenwich Mean Time (8 am local time in Gloucester […]