“The IO infrastructure of TSUBAME3.0 combines fast in-node NVMe SSDs and a large, fast, Lustre-based system from DDN. The 15.9PB Lustre* parallel file system, composed of three of DDN’s high-end ES14KX storage appliances, is rated at a peak performance of 150GB/s. The TSUBAME collaboration represents an evolutionary branch of HPC that could well develop into the dominant HPC paradigm at about the time the most advanced supercomputing nations and consortia achieve Exascale computing.”
In this video, Dr Tim Stitt from the Earlham Institute describes why moving their HPC workload to Iceland made economic sense. Through the Verne Global datacenter, the Earlham Institute will have access to one of the world’s most reliable power grids producing 100% geothermal and hydro-electric renewable energy. As EI’s HPC analysis requirements continue to grow, Verne Global will enable the institute to save up to 70% in energy costs (based on 14p to 4p KWH rate and with no additional power for cooling, significantly benefiting the organization in their advanced genomics and bioinformatics research of living systems.
Today Dutch startup Asperitas rolled out Immersed Computing cooling technology for datacenters. “The company’s first market ready solution, the AIC24, ‘the first water-cooled oil-immersion system which relies on natural convection for circulation of the dielectric liquid.’ This results in a fully self-contained and Plug and Play modular system. The AIC24 needs far less infrastructure than any other liquid installation, saving energy and costs on all levels of datacentre operations. The AIC24 is the most sustainable solution available for IT environments today. Ensuring the highest possible efficiency in availability, energy reduction and reuse, while increasing capacity. Greatly improving density, while saving energy at the same time.”
“TSUBAME3.0 is expected to deliver more than two times the performance of its predecessor, TSUBAME2.5,” writes Marc Hamilton from Nvidia. “It will use Pascal-based Tesla P100 GPUs, which are nearly three times as efficient as their predecessors, to reach an expected 12.2 petaflops of double precision performance. That would rank it among the world’s 10 fastest systems according to the latest TOP500 list, released in November. TSUBAME3.0 will excel in AI computation, expected to deliver more than 47 PFLOPS of AI horsepower. When operated concurrently with TSUBAME2.5, it is expected to deliver 64.3 PFLOPS, making it Japan’s highest performing AI supercomputer.”
“Servers today have hundreds of knobs that can be tuned for performance and energy efficiency. While some of these knobs can have a dramatic effect on these metrics, manually tuning them is a tedious task. It is very labor intensive, it requires a lot of expertise, and the tuned settings are only relevant for the hardware and software that were used in the tuning process. In addition to that, manual tuning can’t take advantage of application phases that may each require different settings. In this presentation, we will talk about the concept of dynamic tuning and its advantages. We will also demo how to improve performance using manual tuning as well as dynamic tuning using DatArcs Optimizer.”
Exxon Mobil and NCSA Achieve New Levels of Scalability on complex Oil & Gas Reservoir Simulation Models
“This breakthrough has unlocked new potential for ExxonMobil’s geoscientists and engineers to make more informed and timely decisions on the development and management of oil and gas reservoirs,” said Tom Schuessler, president of ExxonMobil Upstream Research Company. “As our industry looks for cost-effective and environmentally responsible ways to find and develop oil and gas fields, we rely on this type of technology to model the complex processes that govern the flow of oil, water and gas in various reservoirs.”
SC17 has issued its Call for Panel Sessions. The conference takes place Nov. 12-17 in Denver. “As in past years, panels at SC17 will be some of the most heavily attended events of the Conference. Panels will bring together the key thinkers and producers in the field to consider in a lively and rapid-fire context some of the key questions challenging high performance computing, networking, storage and associated analysis technologies for the foreseeable future.”
“In recent years, major breakthroughs were achieved in different fields using deep learning. From image segmentation, speech recognition or self-driving cars, deep learning is everywhere. Performance of image classification, segmentation, localization have reached levels not seen before thanks to GPUs and large scale GPU-based deployments, leading deep learning to be a first class HPC workload.”
Today ISC 2017 announced that it’s Distinguished Talk series will focus on Data Analytics in manufacturing and scientific applications. One of the Distinguished Talks will be given by Dr. Sabine Jeschke from the Cybernetics Lab at the RWTH Aachen University on the topic of, “Robots in Crowds – Robots and Clouds.” Jeschke’s presentation will be followed by one from physicist Kerstin Tackmann, from the German Electron Synchrotron (DESY) research center, who will discuss big data and machine learning techniques used for the ATLAS experiment at the Large Hadron Collider.
“As with all new technology, developers will have to create processes in order to modernize applications to take advantage of any new feature. Rather than randomly trying to improve the performance of an application, it is wise to be very familiar with the application and use available tools to understand bottlenecks and look for areas of improvement.”