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

Unlike Oil and Water, Legacy and Cloud Mix Well

Despite the cloud hype, legacy HPC apps are alive and well. While it may seem like they can’t mix, the process of bursting these applications to the cloud is bringing these staples to cloud table. Avoiding rewrites can efficiently bring immediate HPC cloud benefits to organizations big and small. “Many technologies and solutions now available allow for the functional and highly efficient coordination and connection between legacy applications and the well-known advantages of the cloud.”

Video: Baidu Releases Fast Allreduce Library for Deep Learning

In this video, from Andrew Gibiansky from Baidu describes baidu-allreduce, a newly released C library that enables faster training of neural network models across many GPUs. The library demonstrates the allreduce algorithm, which you can embed into any MPI-enabled application. “”Neural networks have grown in scale over the past several years, and training can require a massive amount of data and computational resources. To provide the required amount of compute power, we scale models to dozens of GPUs using a technique common in high-performance computing but underused in deep learning.”

Fayetteville State University Leverages Bright Cluster Manager for AI Research

Fayetteville State University is using Bright Cluster Manager for ground-breaking Artificial Intelligence research. Bright Cluster Manager enables the University to deploy complete clusters over bare metal and manage them effectively. Managing the hardware, operating system, HPC software, and its users is done with an elegant easy-to-use graphical user interface. With Bright Cluster Manager, IT/system administrators can quickly get clusters up and running and keep them running reliably throughout their lifecycle.

PRACE Offers Supercomputing 101 Course

PRACE is offering an online Supercomputing 101 course through the Future Learn program. This free online course will introduce you to what supercomputers are, how they are used and how we can exploit their full computational potential to make scientific breakthroughs. “Using supercomputers, we can now conduct virtual experiments that are impossible in the real world.”

Radio Free HPC Looks at Eclipse Simulations

In this podcast, the Radio Free HPC team discusses the eclipse and how researchers are using supercomputers to simulate the corona of the sun at eclipse time. After that, look at the the top Technology stories in our Catch of the Week:

Video: What is Wrong with Convolutional Neural Nets?

Geoffrey Hinton from the University of Toronto gave this talk at the Vector Institute. “What is Wrong with ‘standard’ Convolutional Neural Nets? They have too few levels of structure: Neurons, Layers, and Whole Nets. We need to group neurons in each layer in ‘capsules’ that do a lot of internal computation and then output a compact result.”

Video: Customers Leverage HPE & Intel Alliance for HPC

“Meeting the Enterprise Challenges of HPC System developers and users face obstacles deploying complex new HPC technologies, such as: energy efficiency, reliability and resiliency requirements, or developing software to exploit HPC hardware. All can delay HPC adoption. But the HPC Alliance will help. HPE and Intel will collaborate with you on workstream integration, solution sizing and software-to-hardware integration. We will help, whether you seek ease of everything, to mix workloads, or have fast-growing, increasingly complex systems.”

NeSI in New Zealand Installs Pair of Cray Supercomputers

The New Zealand Science Infrastructure (NeSI) is commissioning a new HPC system that will be colocated at two facilities. “The new systems, provide a step change in power to NeSI’s existing services, including a Cray XC50 Supercomputer and a Cray CS400 cluster High Performance Computer, both sharing the same high performance and offline storage systems.”

Researchers Use TACC, SDSC and NASA Supercomputers to Forecast Corona of the Sun

Predictive Sciences ran a large-scale simulation of the Sun’s surface in preparation for a prediction of what the solar corona will look like during the eclipse. “The Solar eclipse allows us to see levels of the solar corona not possible even with the most powerful telescopes and spacecraft,” said Niall Gaffney, a former Hubble scientist and director of Data Intensive Computing at the Texas Advanced Computing Center. “It also gives high performance computing researchers who model high energy plasmas the unique ability to test our understanding of magnetohydrodynamics at a scale and environment not possible anywhere else.”

Video: Deep Learning on Azure with GPUs

In this video, you’ll learn how to start submitting deep neural network (DNN) training jobs in Azure by using Azure Batch to schedule the jobs to your GPU compute clusters. “Previously, few people had access to the computing power for these scenarios. With Azure Batch, that power is available to you when you need it.”