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


Call for Submissions: 2018 Perth HPC Conference

The 2018 Perth HPC Conference has announced their Call for Submissions. Hosted by the Pawsey Supercomputing Centre, The event takes place August 28-29 in Perth, Australia. “The second annual HPCAIAC Perth Conference will bring industry luminaries and leading subject matter experts together from all over the world to explore major and emerging domains and share in-depth insights on AI, Data Sciences, HPC, Deep/Machine Learning, Visualization and more.”

Video: Demystifying Parallel and Distributed Deep Learning

Torsten Hoefler from (ETH) Zürich gave this talk at the 2018 Swiss HPC Conference. “Deep Neural Networks (DNNs) are becoming an important tool in modern computing applications. Accelerating their training is a major challenge and techniques range from distributed algorithms to low-level circuit design. In this talk, we describe the problem from a theoretical perspective, followed by approaches for its parallelization.”

Mellanox powers NVIDIA DGX-2 AI Supercomputer

Today Mellanox announced that the company’s InfiniBand and Ethernet solutions have been chosen to accelerate the new NVIDIA DGX-2 artificial intelligence system. DGX-2 is the first 2 Petaflop system that combines sixteen GPUs and eight Mellanox ConnectX adapters, supporting both EDR InfiniBand and 100 gigabit Ethernet connectivity. The technological advantages of the Mellanox adapters with smart acceleration engines enable the highest performance for AI and Deep Learning applications.

Video: Liqid Teams with Inspur at GTC for Composable Infrastructure

In this video from GTC 2018, Dolly Wu from Inspur and Marius Tudor from Liquid describe how the two companies are collaborating on Composable Infrastructure for AI and Deep Learning workloads. “AI and deep learning applications will determine the direction of next-generation infrastructure design, and we believe dynamically composing GPUs will be central to these emerging platforms,” said Dolly Wu, GM and VP Inspur Systems.

Jobs of the Week: Software/Hardware Architects for Research Infrastructure at University of Southern Denmark

The eScience Center at the University of Southern Denmark (SDU) is seeking Software/Hardware Architects for Research Infrastructure. “The eScience Center is now expanding its staff in all its core areas and expects to fill 6 or more positions. We address stimulating and interesting technological challenges, and offer a research-like environment where we encourage the study of innovative solutions. We firmly believe in open-source and web technology as means to reach our goals.

NVMe Over Fabrics High performance SSDs Networked for Composable Infrastructure

Rob Davis from Mellanox gave this talk at the 2018 OCP Summit. “There is a new very high performance open source SSD interfaced called NVMe over Fabrics now available to expand the capabilities of networked storage solutions. It is an extension of the local NVMe SSD interface developed a few years ago driven by the need for a faster interface for SSDs. Similar to the way native disk drive SCSI protocol was networked with Fibre Channel 20 years ago, this technology enables NVMe SSDs to be networked and shared with their native protocol. By utilizes ultra-low latency RDMA technology to achieve data sharing across a network without sacrificing the local performance characteristics of NVMe SSDs, true composable infrastructure is now possible.”

Video: Powering the Road to National HPC Leadership

Jack Wells from ORNL gave this talk at the 2018 OpenPOWER Summit. “The Summit supercomputer coming to Oak Ridge is the next leap in leadership-class computing systems for open science. Summit will have a hybrid architecture, and each node will contain multiple IBM POWER9 CPUs and NVIDIA Volta GPUs all connected together with NVIDIA’s high-speed NVLink. Each node will have over half a terabyte of coherent memory (high bandwidth memory + DDR4) addressable by all CPUs and GPUs plus 800GB of non-volatile RAM that can be used as a burst buffer or as extended memory.”

Video: Gen-Z High-Performance Interconnect for the Data-Centric Future

Greg Casey from Dell gave this talk at the 2018 OCP Summit. “Gen-Z is different. It is a high-bandwidth, low-latency fabric with separate media and memory controllers that can be realized inside or beyond traditional chassis limits. It treats all components as memory (so-called memory-semantic communications), and it moves data between them with minimal overhead and latency. It thus takes full advantage of emerging persistent memory (memory accessed over the data bus at memory speeds). It can also handle other compute elements, such as GPUs, FPGAs, and ASIC or coprocessor-based accelerators.”

What a 200G HDR InfiniBand Solution Means for Today’s Advanced Data Centers

The world of today’s HPC computing is driven by the ever-increasing generation and consumption of digital information. And the ability to analyze this rapidly growing pool of data, and extrapolate meaningful insights, gives modern businesses a competitive edge. Download the full report, “Introducing 200G HDR InfiniBand Solutions,” to learn about how Mellanox Technologies end-to-end 200G HDR InfiniBand solution is helping enable the next generation of data centers.

Agenda Posted for GlobusWorld 2018 in Chicago

The good folks at GlobusWorld 2018 have posted their speaker agenda. The meeting takes place April 25-26 in Chicago. “Our theme for 2018 is Connecting your Research Universe. When it’s easy for researchers to move and share data across any location (e.g. campus clusters, XSEDE systems, lab servers, scientific instruments, archival storage, commercial clouds), they can collaborate more successfully and get to results faster. Globus now makes this possible, with connectivity to virtually any storage location via a single, easy to use interface. It has never been simpler to connect the universe of research data to the researchers who need it.”