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

A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator Servers

“MeteoSwiss, the Swiss national weather forecast institute, has selected densely populated accelerator servers as their primary system to compute weather forecast simulation. Servers with multiple accelerator devices that are primarily connected by a PCI-Express (PCIe) network achieve a significantly higher energy efficiency. Memory transfers between accelerators in such a system are subjected to PCIe arbitration policies. In this paper, we study the impact of PCIe topology and develop a congestion-aware performance model for PCIe communication. We present an algorithm for computing congestion factors of every communication in a congestion graph that characterizes the dynamic usage of network resources by an application.”

Panel Discussion: The Exascale Era

In this video from Switzerland HPC Conference, Rich Brueckner from insideHPC moderates a panel discussion on Exascale Computing. “The Exascale Computing Project in the USA is tasked with developing a set of advanced supercomputers with 50x better performance than today’s fastest machines on real applications. This panel discussion will look at the challenges, gaps, and probable pathways forward in this monumental endeavor.”


Gilad Shainer, HPC Advisory Council
Jeffrey Stuecheli, IBM
DK Panda, Ohio State University
Torsten Hoefler, ETH Zurich
Rich Graham, Mellanox

A Fresh Look at HPC from Huawei Enterprise

“High performance computing is rapidly finding new uses in many applications and businesses, enabling the creation of disruptive products and services. Huawei, a global leader in information and communication technologies, brings a broad spectrum of innovative solutions to HPC. This talk examines Huawei’s world class HPC solutions and explores creative new ways to solve HPC problems.”

dCUDA: Distributed GPU Computing with Hardware Overlap

“Over the last decade, CUDA and the underlying GPU hardware architecture have continuously gained popularity in various high-performance computing application domains such as climate modeling, computational chemistry, or machine learning. Despite this popularity, we lack a single coherent programming model for GPU clusters. We therefore introduce the dCUDA programming model, which implements device-side remote memory access.”

High-Performance and Scalable Designs of Programming Models for Exascale Systems

“This talk will focus on challenges in designing programming models and runtime environments for Exascale systems with millions of processors and accelerators to support various programming models. We will focus on MPI+X (PGAS – OpenSHMEM/UPC/CAF/UPC++, OpenMP, and CUDA) programming models by taking into account support for multi-core systems (KNL and OpenPower), high-performance networks, GPGPUs (including GPUDirect RDMA), and energy-awareness.”

Video: Advancing Open Fabrics Interfaces

Sean Hefty from Intel presented this talk at the OpenFabrics Workshop. “With its initial release two years ago, libfabric advanced the state of fabric software interfaces. One of the promises of OFI was extensibility: adapting to increased demands of fabric services from applications. This session explores the first major enhancements to the libfabric API in response to user demands and learnings.”

Visionary Perspective: Foundations of Cognitive Computing

Costas Bekas from IBM Research Zurich presented this talk at the Switzerland HPC Conference. “IBM Research builds applications that enable humans to collaborate with powerful AI technologies to discover, analyze and tackle the world’s greatest challenges. Humans are on the cusp of augmenting their lives in extraordinary ways with AI. At IBM Research Labs around the globe, we envision and develop next-generation systems that work side-by side with humans, accelerating our ability to create, learn, make decisions and think.”

Cowboy Supercomputer Powers Research at Oklahoma State

In this video, Dana Brunson from Oklahoma State describes the mission of the Oklahoma High Performance Computing Center. Formed in 2007, the HPCC facilitates computational and data-intensive research across a wide variety of disciplines by providing students, faculty and staff with cyberinfrastructure resources, cloud services, education and training, bioinformatics assistance, proposal support and collaboration.

Managing Node Configuration with 1000s of Nodes

Ira Weiny from Intel presented this talk at the OpenFabrics Workshop. “Individual node configuration when managing 1000s or 10s of thousands of nodes in a cluster can be a daunting challenge. Two key daemons are now part of the rdma-core package which aid the management of individual nodes in a large fabric: IBACM and rdma-ndd.”

Dr. Robert Voigt on Educating Computational Scientists

In this video from KAUST Live, Dr. Robert Voigt discusses his recent keynote at the HPC Saudi Conference on the topic of Educating Computational Scientists. “This talk will provide a historical perspective on the challenges of educating computational scientists based on my personal involvement over a number of years. Three decidedly different activities will be drawn on to indicate how one can successfully approach the challenge.”