Archive for the ‘supercomputing’ category: Page 72
Sep 20, 2017
China Upgrading Milky Way 2 Supercomputer to 95 Petaflops
Posted by Roman Mednitzer in category: supercomputing
We have some breaking news from the IHPC Forum in Guangzhou today. Researchers in China are busy upgrading the MilkyWay 2 (Tianhe-2) system to nearly 95 Petaflops (peak). This should nearly double the performance of the system, which is currently ranked at #2 on TOP500 with 33.86 Petaflops on the Linpack benchmark. The upgraded system, dubbed Tianhe −2A, should be completed in the coming months.
Details about the system upgrade were presented at the conference opening session. While the current system derives much of its performance from Intel Knights Corner co-processors, the new system swaps these PCI devices out for custom-made 4-way MATRIX-200o boards, with each chip providing 2.46 Teraflops of peak performance.
Continue reading “China Upgrading Milky Way 2 Supercomputer to 95 Petaflops” »
Sep 9, 2017
EU Funds 20 Million Euro ARM+FPGA Exascale Project
Posted by Roman Mednitzer in categories: energy, supercomputing
At the Barcelona Supercomputer Centre on Wednesday (Sept. 6), 16 partners gathered to launch the EuroEXA project, which invests €20 million over three-and-a-half years into exascale-focused research and development.
Led by the Horizon 2020 program, EuroEXA picks up the banner of a triad of partner projects — ExaNeSt, EcoScale and ExaNoDe — building on their work to develop a complete HPC system based on ARM Cortex processors and Xilinx Ultrascale FPGAs. The goal is to deploy an energy-efficient petaflops system by 2020 and lay a path to achieve exascale capability in the 2022–23 timeframe.
All told, the European Commission is planning a €50 million investment for the EuroEXA group of projects, spanning “research, innovation and action across applications, system software, hardware, networking, storage, liquid cooling and data centre technologies.”
Continue reading “EU Funds 20 Million Euro ARM+FPGA Exascale Project” »
Aug 25, 2017
Major leap towards storing data at the molecular level
Posted by Klaus Baldauf in categories: chemistry, mobile phones, supercomputing
From smartphones to supercomputers, the growing need for smaller and more energy efficient devices has made higher density data storage one of the most important technological quests.
Now scientists at the University of Manchester have proved that storing data with a class of molecules known as single-molecule magnets is more feasible than previously thought.
The research, led by Dr David Mills and Dr Nicholas Chilton, from the School of Chemistry, is being published in Nature. It shows that magnetic hysteresis, a memory effect that is a prerequisite of any data storage, is possible in individual molecules at −213 °C. This is tantalisingly close to the temperature of liquid nitrogen (−196 °C).
Continue reading “Major leap towards storing data at the molecular level” »
Aug 2, 2017
AMD reveals PetaFLOP supercomputer in a single rack
Posted by Klaus Baldauf in categories: robotics/AI, supercomputing
Yesterday, AMD revealed the Project 47 supercomputer was powered by 20 AMD EPYC 7601 processors and 80 Radeon Instinct GPUs. It is a petaFLOP supercomputer in a rack. Other hardware included 10TB of Samsung memory and 20 Mellanox 100G cards (and 1 switch). Project 47 is capable of 1 PetaFLOP of single-precision compute performance or 2 PetaFLOPS of half-precision.
Project 47 is built around the Inventec P47. The P47 is a 2U parallel computing platform designed for graphics virtualization and machine intelligence applications. A single rack of Inventec P47 systems is all that was necessary to achieve 1 PetaFLOP, and it does so while producing 30 GigaFLOPS/Watt, which AMD claims is 25% more efficient than some other competing supercomputing platforms. A petaFLOP system uses 33,333 watts. A thousand of PetaFLOP racks would use 33.3 MW and have an exaFLOP.
Thanks to its 32-core / 64-thread EPYC processors and Radeon Vega GPUs, which feature 4,096 stream processors each, AMD also claims that Project 47 rack has more cores/threads, compute units, I/O lanes and memory channels in use simultaneously than in any other similarly configured system.
Continue reading “AMD reveals PetaFLOP supercomputer in a single rack” »
Jul 29, 2017
Edited Humans, Creating A Universe With A Supercomputer & All Is One At The Same Time
Posted by Dave Holt in categories: bioengineering, biotech/medical, supercomputing
Welcome to Mr Futurist’s first weekly podcast where I discuss what going on in emerging science and technology. Scientists have successfully edited the first human embryo in the U.S. using CRISPR. CRISPR is a gene editing technique that can modify any region of the genome of any species with high precision accuracy. Modifying a species to have certain characteristics or traits. If you’re curious as to what CRISPR is, I have added a link below to an excellent video from Futurist that explains what CRISPR is and what it can be used for, all in 60 seconds. It’s worth a minute of your time.
Jul 21, 2017
World dominance in three steps: China sets out road map to lead in artificial intelligence by 2030
Posted by Derick Lee in categories: finance, government, robotics/AI, supercomputing
Government finance will lead the way in AI research, including the development of supercomputers, and high performance semiconductor chips, software and the hiring of key talent to lead the field, China’s science and technology minister Wan Gang said in March during the country’s parliamentary meeting.
The Chinese government’s July 8 plan aims to keep pace with AI technology by 2020, make major breakthroughs by 2025, and lead the world in AI by 2030.
PUBLISHED : Friday, 21 July, 2017, 1:28pm.
Jun 10, 2017
Are We Building Artificial Brains And Uploading Minds To The Cloud Right Now?
Posted by Dave Holt in categories: robotics/AI, supercomputing
https://youtube.com/watch?v=amwyBmWxESA
When people post their emotional responses to social media and through their free email account(s), they are loading their human personal emotional responses, judgments, and biases into a large computer and cloud database? Everything we post and respond to is data somewhere. The truth is, hundreds of millions of people around the planet do this every day, 24 hours a day, seven days a week.
Are we uploading our brains to a cloud on a supercomputer and evolving into an artificially intelligent machine? This question and more…
Continue reading “Are We Building Artificial Brains And Uploading Minds To The Cloud Right Now?” »
Jun 2, 2017
A Net Neutrality Nightmare? / Part II (Future A to Z)
Posted by Johnny Boston in categories: futurism, information science, internet, journalism, law, media & arts, software, strategy, supercomputing
The recent efforts to remove Net Neutrality have given many a sense of impending doom we are soon to face. What happens to an Internet without Net Neutrality? Advocates have a vision of the possible results — and it is quite the nightmare! In this segment of Future A to Z, The Galactic Public Archives takes a cheeky, yet compelling perspective on the issue.
May 17, 2017
Google announces a powerful new AI chip and supercomputer
Posted by Shane Hinshaw in categories: robotics/AI, supercomputing
The new chip and a cloud-based machine-learning supercomputer will help Google establish itself as an AI-focused hardware maker.