Over at the Nvidia Blog, George Millington writes that, the fourth consecutive year, the Nvidia Tesla Accelerated Computing Platform helped set new milestones in the Asia Student Supercomputer Challenge, the world’s largest supercomputer competition.
“Single Root I/O Virtualization (SR-IOV) technology has been steadily gaining momentum for high-performance interconnects such as InfiniBand. SR-IOV can deliver near native performance but lacks locality-aware communication support. This talk presents an efficient approach to build HPC clouds based on MVAPICH2 over OpenStack with SR-IOV. We discuss the high-performance design of virtual machine-aware MVAPICH2 library over OpenStack-based HPC Clouds with SR-IOV. A comprehensive performance evaluation with micro-benchmarks and HPC applications has been conducted on an experimental OpenStack-based HPC cloud and Amazon EC2. The evaluation results show that our design can deliver near bare-metal performance.”
In this video from the OpenPOWER Summit, Dr. Mike Ashworth from the STFC Hartree Centre presents: Driving Innovation at the Hartree Centre with the OpenPOWER Architecture. “The STFC Hartree Centre has been established with a focus on economic impact through our collaborations with UK industry, offering access to HPC and data intensive systems, code development expertise and collaborative research opportunities. Through a major partnership with IBM we have been porting and evaluating our key application codes on the OpenPOWER architecture seeking to leverage performance and productivity gains for the benefit of our key customers. We will describe progress in porting and optimization of applications from a range of science areas.”
Designing materials atom-by-atom has long been a science fiction dream. Georg Schusteritsch and Chris Pickard of the University of Cambridge are bringing science fiction one step closer to reality using the UK National Supercomputing Facility, ARCHER to reveal the interfaces forming within and between materials. “We have developed a general first-principles approach to predict the crystal structure of interfaces in materials, a technique that represents a major step towards computationally developing materials with specially designed interfaces.”
Parallel programming software developed at the Barcelona Supercomputing Center is being used as part of a successful kickstarter campaign for a maker board with 10 times the power of the Rasberry PI 3. “Developed at BSC, the OmpSs parallel programming model is used on the new UDOO X86 board, which reached its Kickstarter funding target of €100,000 in under seven hours. UDOO X86 combines a powerful maker board and an Arduino 101-compatible platform, all embedded on the same board. It can be used for a range of applications, such as gaming, video streaming, graphic design editing, Internet of Things applications, or as a toolbox for makers.”
Over at the Women in HPC Blog, Daniel Holmes from EPCC writes that the EuroMPI Conference is partnering with Women in HPC to increase diversity in high performance computing.
“Fortissimo will make advanced simulation more easily accessible, particularly to SMEs, through the realization of a “one-stop shop” where hardware, expertise, applications, visualization and tools will be easily available and affordable on a pay-per-use basis. In doing this, it will create and demonstrate a sustainable commercial ecosystem where actors at all levels in the value chain can realize sufficient commercial benefit to enable that ecosystem to persist independently of EU funding and continue to provide affordable services to manufacturing industry, particularly SMEs.”
Today the Information Technology and Innovation Foundation (ITIF) published a new report that urges U.S. policymakers to take decisive steps to ensure the United States continues to be a world leader in high-performance computing. “While America is still the world leader, other nations are gaining on us, so the U.S. cannot afford to rest on its laurels. It is important for policymakers to build on efforts the Obama administration has undertaken to ensure the U.S. does not get out paced.”
“Cycle Computing software leverages cloud resources to make computation in the cloud productive at any scale, by orchestrating workflows, managing data, balancing cloud options, and enabling users in a secure, controlled way. Our software works with public cloud providers like Amazon Web Services, Google Cloud Platform, and Microsoft Azure, as well as with internal and private cloud environments.”
In this TACC podcast, Joe Stubbs from the Texas Advanced Computing Centter describes potential benefits to scientists of open container platform Docker in supporting reproducibility, NSF-funded Agave API. “As more scientists share not only their results but their data and code, Docker is helping them reproduce the computational analysis behind the results. What’s more, Docker is one of the main tools used in the Agave API platform, a platform-as-a-service solution for hybrid cloud computing developed at TACC and funded in part by the National Science Foundation.”