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

OpenPOWER Developer Congress Event to Focus on Machine Learning

Today IBM announced that the first annual OpenPOWER Foundation Developer Congress will take place May 22-25 in San Francisco. With a focus on Machine Learning, the conference will focus on continuing to foster the collaboration within the foundation to satisfy the performance demands of today’s computing market.

Baidu Deep Learning Service adds Latest NVIDIA Pascal GPUs

“Baidu and NVIDIA are long-time partners in advancing the state of the art in AI,” said Ian Buck, general manager of Accelerated Computing at NVIDIA. “Baidu understands that enterprises need GPU computing to process the massive volumes of data needed for deep learning. Through Baidu Cloud, companies can quickly convert data into insights that lead to breakthrough products and services.”

Podcast: How AI Can Improve the Diagnosis and Treatment of Diseases

In this AI Podcast, Mark Michalski from the Massachusetts General Hospital Center for Clinical Data Science discusses how AI is being used to advance medicine. “Medicine — particularly radiology and pathology — have become more data-driven. The Massachusetts General Hospital Center for Clinical Data Science — led by Mark Michalski — promises to accelerate that, using AI technologies to spot patterns that can improve the detection, diagnosis and treatment of diseases.”

Deep Learning on the SaturnV Cluster

“The basic idea of deep learning is to automatically learn to represent data in multiple layers of increasing abstraction, thus helping to discover intricate structure in large datasets. NVIDIA has invested in SaturnV, a large GPU-accelerated cluster, (#28 on the November 2016 Top500 list) to support internal machine learning projects. After an introduction to deep learning on GPUs, we will address a selection of open questions programmers and users may face when using deep learning for their work on these clusters.”

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

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.”

Tesla P100 GPUs Speed Cloud-Based Deep Learning on the Nimbix Cloud

“Nimbix has tremendous experience in GPU cloud computing, going all the way back to NVIDIA’s Fermi architecture,” said Steve Hebert, CEO of Nimbix. “We are looking forward to accelerating deep learning and analytics applications for customers seeking the latest generation GPU technology available in a public cloud.”

GPU Hackathon Coming to Spain End of May

Appentra and CESGA are organizing a GPU Hackathon in Spain. The event takes place May 29 – June 1 at the Galicia Supercomputing Center in Santiago de Compostela. The event is free, limited in capacity and in Spanish, with priority for national participants.

NVIDIA Tesla P100 GPU Speeds AI Workloads in the IBM Cloud

Today IBM announced that it is the first major cloud provider to make the Nvidia Tesla P100 GPU accelerator available globally on the cloud. “As the AI era takes hold, demand continues to surge for our GPU-accelerated computing platform in the cloud,” said Ian Buck, general manager, Accelerated Computing, NVIDIA. “These new IBM Cloud offerings will provide users with near-instant access to the most powerful GPU technologies to date – enabling them to create applications to address complex problems that were once unsolvable.”