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

Archives for September 2011

Lustre 2.1 is Out, First Major Release with OpenSFS Hosting the Master Codeline

Today Whamcloud announced the release of Lustre 2.1 with better control over virtualized storage and wider platform support. Here is the lowdown: Lustre 2.1 available immediately on Whamcloud servers Release includes interesting user-requests features including bigger drives and RHEL support Community contribution rules have been simplified This is the first full release of Lustre with […]

Video: The Real Future of Cloud Computing

In this video, Jeff Broughton from NERSC presents: The Real Future of Cloud Computing. This video is well-worth a look — Broughton is quite candid in this talk about the limitations of cloud computing as discovered in the Magellan project. This presentation was recorded at the HPC User Forum in San Diego on Sept. 8, […]

Video: November HP-CAST User Group to Focus on HPC, Cloud, and Exascale

In this video, Ed Turkel discusses HP-CAST (HP’s Consortium for Advanced Scientific and Technical Computing Users Group meeting), which is coming up Nov. 11-12 in Seattle just prior to SC11. I attended the recent HP-CAST in Hamburg and the technical sessions were terrific. In fact, it made me a little nostalgic for the Sun HPC […]

AWS Spot Instances Slash the Cost of Scientific Computing

Amazon’s Jeff Barr writes that AWS EC2 Spot Instances were tailor-made for the HPC needs of the price-sensitive scientific community. The scientific community was quick to recognize that their compute-intensive, batch workloads (often known as HPC or Big Data) were a perfect fit for EC2 Spot Instances. These AWS customers have seen cost savings of […]

Russia Enters the Exascale Cold War – Targets 2020

Clipped from: (share this clip) According to a CNews report, Russia researchers have drafted plans to build an exascale supercomputer by 2020, joining a list of countries with similar aspirations including China, Japan, and the United States. Sources close to the project reportedly estimated the project to up to $1.8 Billion USD. Russia recently […]

Additional Presentations Posted from European Lustre Workshop

The first EOFS European Lustre Workshop happened this week in Paris. With 63 attendees representing 27 organizations, the event was reportedly as big as LUG ’07. EOFS has posted the following presentations from the Workshop: An update on both the 2.1 community release and also WC Lustre releases Peter Jones, Whamcloud Lessons learned from Lustre […]

Video: Microsoft – Datacenter Efficiency – Optimizing Performance and Power

In this video, Ryan Waite, GM of HPC for Microsoft presents: Datacenter Efficiency – Optimizing Performance and Power. Recorded at the HPC User Forum in San Diego on Sept. 7, 2011. There’s still time to register for the HPC User Forum series coming to Europe next week: At CEA in France, Oct. 3-4. Agenda and […]

Intel's OpenCL Autovectorization Boosts Performance without User Intervention

Intel’s Nadav Rotem writes that the company’s newly released OpenCL SDK version 1.5 features one improvement that is very important but not always visible to the user: the new Implicit CPU Vectorization module. What are the benefits of using the implicit CPU vectorization module? SIMD instructions expose a high-level of parallelism and are used to […]

Podcast: jclouds' Adrian Cole on Next-Gen Cloud Infrastructure

In this podcast, Rich Brueckner of interviews Adrian Cole, Chief Evangelist at Cloudsoft and founder at jclouds is an open source library that helps you get started in the cloud and reuse your java and clojure development skills. Our api allows you freedom to use portable abstractions or cloud-specific features. We test support […]

Floyer: Gigabit Ethernet is Not Taking Over the World of HPC

Wikibon CTO David Floyer reports that Gigabit Ethernet is not taking over the world of HPC. If part of an organization is saying that Ethernet should be the only interconnect, make sure they do not have responsibility for performance IO and Performance computing. If a vendor is arguing that Ethernet is the only way to […]