Faster Fabrics Running Against Limits of the Operating System, the Processor, and the I/O Bus
Christopher Lameter from Jump Trading gave this talk at the OpenFabrics Workshop in Austin. “In 2017 we got 100G fabrics, in 2018 200G fabrics and in 2019 it looks like 400G technology may be seeing a considerable amount of adoption. These bandwidth compete with and sometimes are higher than the internal bus speeds of the servers that are connected using these fabrics. I think we need to consider these developments and work on improving fabrics and the associated APIs so that ways to access these features become possible using vendor neutral APIs. It needs to be possible to code in a portable way and not to a vendor specific one.” [READ MORE…]
Featured Stories
E-CAS Project to Explore Clouds for Acceleration of Science
Can the Cloud power ground-breaking research? A new NSF-funded research project aims to provide a deeper understanding of the use of cloud computing in accelerating scientific discoveries. “First announced in 2018, the Exploring Clouds for Acceleration of Science (E-CAS) project has now selected the six research proposals to explore how scientific workflows can leverage advancements in real-time analytics, artificial intelligence, machine learning, accelerated processing hardware, automation in deployment and scaling, and management of serverless applications for a wider range of science.” [READ MORE…]
D-Wave Leap Quantum Cloud Service comes to Europe and Japan
Today D-Wave Systems announced the geographic expansion of its Leap quantum cloud service to 33 countries in Europe and Asia. According to the company, Leap is the only cloud-based service to provide real-time access to a live quantum computer, as well as open- source development tools, interactive demos, educational resources, and knowledge base articles. “The range and robustness of early applications from our customers continues to grow, and customers are starting to see early value in using quantum computing to address real-world business problems.” [READ MORE…]
Exploring the ROI Potential of GPU Supercomputing
The growing prevalence of artificial intelligence and machine learning is putting heightened focus on the quantities of data that organizations have recently accumulated — as well as the value potential in this data. Companies looking to gain a competitive edge in their market are turning to tools like graphic processing units – or GPUs – to ramp up computing power. That’s according to a new white paper from Penguin Computing. [READ MORE…]
Featured Resource
Introducing 200G HDR InfiniBand Solutions
As the first to 40Gb/s, 56Gb/s and 100Gb/s bandwidth, Mellanox has both boosted data center and cloud performance and improved return on investment at a pace that exceeds its own roadmap. To that end, Mellanox has now announced that it is the first company to enable 200Gb/s data speeds with Mellanox Quantum switches, ConnectX-6 adapters, and LinkX cables combining for an end-to end 200G HDR InfiniBand solution in 2018. Download the new report, courtesy of Mellanox Technologies, to lean more about 200G HDR InfiniBand solutions.
Industry Perspectives
CPU, GPU, FGPA, or DSP: Heterogeneous Computing Multiplies the Processing Power
Whether your code will run on industry-standard PCs or is embedded in devices for specific uses, chances are there’s more than one processor that you can utilize. Graphics processors, DSPs and other hardware accelerators often sit idle while CPUs crank away at code better served elsewhere. This sponsored post from Intel highlights the potential of Intel SDK for OpenCL Applications, which can ramp up processing power. [READ MORE…]
Video: Cray Announces First Exascale System
In this video, Cray CEO Pete Ungaro announces Aurora – Argonne National Laboratory’s forthcoming supercomputer and the United States’ first exascale system. Ungaro offers some insight on the technology, what makes exascale performance possible, and why we’re going to need it. “It is an exciting testament to Shasta’s flexible design and unique system and software capabilities, along with our Slingshot interconnect, which will be the foundation for Argonne’s extreme-scale science endeavors and data-centric workloads. Shasta is designed for this transformative exascale era and the convergence of artificial intelligence, analytics and modeling and simulation– all at the same time on the same system — at incredible scale.” [READ MORE…]
Featured from insideBIGDATA
- Risky Business: How Organizations Can Navigate Privacy and Compliance in a GDPR WorldIn this special guest feature, Cory Cowgill, Fusion Risk Management Chief Technology Officer, discusses how enterprise companies should no longer see business continuity and risk management as two separate entities, but marry both practices together. This method helps create a holistic view and enterprise companies get the best of both worlds. It ensures they understand, […]
News from insideBIGDATA
- Building a Data Catalog: A Guide to Planning & Implementing
- Risky Business: How Organizations Can Navigate Privacy and Compliance in a GDPR World
- A ‘Pre-Flight Checklist’ for Machine Learning Training Data
- The Big Data Era: Managing Challenges of Scale, Speed, Personal Information
- SAS Announces $1 Billion Investment in AI at #GTC19
- Unexpected Career Paths That Use Big Data
- GPU Databases On the Rise at NVIDIA #GTC19
Editor’s Choice
Oracle Cloud Speeds HPC & Ai Workloads at GTC 2019
In this video from the GPU Technology Conference, Karan Batta from Oracle describes how the company provides HPC and Machine Learning in the Cloud with Bare Metal speed. ” Oracle Cloud Infrastructure offers wide-ranging support for NVIDIA GPUs, including the high-performance NVIDIA Tesla P100 and V100 GPU instances that provide the highest ratio of CPU cores and RAM per GPU available. With a maximum of 52 physical CPU cores, 8 NVIDIA Volta V100 units per bare metal server, 768 GB of memory, and two 25 Gbps interfaces, these are the most powerful GPU instances on the market.” [READ MORE…]
PASC19 Preview: Brueckner and Dr. Curfman-McInnes to Moderate Exascale Panel Discussion
Today the PASC19 Conference announced that Dr. Lois Curfman McInnes from Argonne and Rich Brueckner from insideHPC will moderate a panel discussion with thought leaders focused on software challenges for Exascale and beyond. “In this session, Lois Curfman McInnes from Argonne National Laboratory and Rich Brueckner from insideHPC will moderate a panel discussion with thought leaders focused on software challenges for Exascale and beyond – mixing “big picture” and technical discussions. McInnes will bring her unique perspective on emerging Exascale software ecosystems to the table, while Brueckner will illustrate the benefits of Exascale to world-wide audiences.” [READ MORE…]
How Mellanox SHARP technology speeds Ai workloads
Mellanox Scalable Hierarchical Aggregation and Reduction Protocol (SHARP) technology improves upon the performance of MPI operations by offloading collective operations from the CPU to the switch network, and by eliminating the need to send data multiple times between endpoints. This innovative approach decreases the amount of data traversing the network as aggregation nodes are reached, and dramatically reduces the MPI operations time. Implementing collective communication algorithms in the network also has additional benefits, such as freeing up valuable CPU resources for computation rather than using them to process communication.” [READ MORE…]
Video: The Game Changing Post-K Supercomputer for HPC, Big Data, and Ai
Satoshi Matsuoka from RIKEN gave this talk at the Rice Oil & Gas Conference. “Rather than to focus on double precision flops that are of lesser utility, rather Post-K, especially its Arm64fx processor and the Tofu-D network is designed to sustain extreme bandwidth on realistic applications including those for oil and gas, such as seismic wave propagation, CFD, as well as structural codes, besting its rivals by several factors in measured performance. Post-K is slated to perform 100 times faster on some key applications c.f. its predecessor, the K-Computer, but also will likely to be the premier big data and AI/ML infrastructure.” [READ MORE…]
Watch Now: the @insideHPC Documentary about Dogs and the People Who Love Them
I would like share this very special film with all my readers who happen to be Dog lovers. “Once there was a Giant” is a short documentary film about Dogs and the Dog Owners who love them. Written and Directed by Richard Brueckner, this film premiered in Portland on Halloween and is now live on YouTube. [READ MORE…]
