MailChimp Developer

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


Offloading Application Segments to Intel Xeon Phi Coprocessors

Offloading to a coprocessor does need to be considered carefully, due to the memory transfer requirements. When the data that is to be worked on resides in the memory of the main system, that data must be transferred to the coprocessor’s memory. The challenge arises because memory is not physically shared between the main system and the coprocessor.
“There are two offload models that the developer must consider when programming an application. The first is the non-shared memory model, and the second is the virtual shared memory model. Both of these models can be used in the same application.”

Scripps Leverages DDN Storage to Research New Medical Treatments

Today DDN announced that The Scripps Research Institute (TSRI), one of the world’s largest independent organizations focusing on biomedical research, has deployed DDN’s end-to-end data management solutions, including high performance SFA7700X file storage automatically tiered to WOS object storage archive, to support fast analysis and cost-effective retention of research data produced by cryo-electron microscopy (Cryo-EM).

3D Printing Survey Provides Insight on First Adopters

While the National Labs are known for their supercomputers, some are also tasked with helping US industry advance digital manufacturing. The 3D printed car and Jeep projects were done to demonstrate Oak Ridge’s Big Area Additive Manufacturing technology, which the lab says could bring a whole new meaning to the phrase “rapid prototyping.” A new report by a 3D printing service called Sculpteo offers some insight into who is using 3D printing. They surveyed 1,000 respondents from 19 different industry online from late January to late March 2016.

Nor-Tech & ESI Partner on Affordable Clusters for OpenFOAM

“We saw a disconnect in the industry between rapidly growing organizations that could really benefit from HPC clusters and the solutions that were on the market,” said Nor-Tech President and CEO David Bollig. “Our goal was to develop a powerful, scalable cluster that was in itself affordable, but also didn’t require hefty, recurring software licensing fees.”

Video: HPE Apollo 6500 Takes GPU Density to the Next Level

In this video from ISC 2016, Greg Schmidt from Hewlett Packard Enterprise describes the new Apollo 6500 server. With up to eight high performance NVIDIA GPUs designed for maximum transfer bandwidth, the HPE Apollo 6500 is purpose-built for HPC and deep learning applications. Its high ratio of GPUs to CPUs, dense 4U form factor and efficient design enable organizations to run deep learning recommendation algorithms faster and more efficiently, significantly reducing model training time and accelerating the delivery of real-time results, all while controlling costs.

Slidecast: Rethinking Server Power Architecture in the Post-Silicon World

In this slidecast, Alexander Lidow from EPC describes how the company is leading a technological revolution with Gallium Nitride (GaN). More efficient than silicon as a basis for electronics, GaN could save huge amounts of energy in the datacenter and has the potential to fuel the computer industry beyond Moore’s Law. “Due to its superior switching speeds and smaller footprint, Texas Instruments is working with EPC to build a simpler topology that achieves better efficiency with smaller footprints and significantly lower cost.”

Cavium Rolls Out ThunderX Servers with GIGABYTE Technology

Today GIGABYTE Technology and Cavium announced a new set of servers built on the industry-leading ThunderX family of workload-optimized ARM server SoCs. According to Cavium, the collaboration brings the world’s most powerful 64-bit ARM-based servers to market to address increasingly demanding application and workload requirements.

Spectra BlackPearl Offers Direct Archive to AWS

Private cloud storage enabled by BlackPearl provides the best option for balancing performance and cost; however, some customer workflows benefit greatly from incorporating a public cloud storage infrastructure,” said Matt Starr, Spectra Logic’s CTO. “Public cloud works best when a disaster recovery copy is needed; when wide data distribution is called for; to create geographic dispersion of data; and when other cloud-based features can be leveraged, such as automatic transcoding. The new Spectra hybrid storage ecosystem means that Spectra customers don’t have to choose between public and private cloud; BlackPearl brings them together seamlessly.”

Intel® Xeon Phi™ Processor—Highly Parallel Computing Engine for HPC

For decades, Intel has been enabling insight and discovery through its technologies and contributions to parallel computing and High Performance Computing (HPC). Central to the company’s most recent work in HPC is a new design philosophy for clusters and supercomputers called Intel® Scalable System Framework (Intel® SSF), an approach designed to enable sustained, balanced performance as the community pushes towards the Exascale era.

SC16 Exhibits Already Breaking Records

While SC16 isn’t until November, the conference is already gaining momentum. There is already a record 188 organizations signed up to exhibit on the industry floor.