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


The true cost of AI innovation

“As the world’s attention has shifted to climate change, the field of AI is beginning to take note of its carbon cost. Research done at the Allen Institute for AI by Roy Schwartz et al. raises the question of whether efficiency, alongside accuracy, should become an important factor in AI research, and suggests that AI scientists ought to deliberate if the massive computational power needed for expensive processing of models, colossal amounts of training data, or huge numbers of experiments is justified by the degree of improvement in accuracy.”

Fugaku Supercomputer joins fight against COVID-19

Today RIKEN in Japan announced that the partially finished Fugaku supercomputer will be made available for research projects aimed to combat COVID-19. The installation of the new supercomputer began in December 2019, and it is scheduled to go into full-fledged open use in 2021. “To combat the global pandemic of the COVID-19 virus, we will rapidly provide access to the capabilities of Fugaku, leapfrogging its preparation, to accelerate the scientific process of diagnosis, treatment, as well as general prevention of infection spread, to contribute to the early termination of the pandemic.”

Video: Profiling Python Workloads with Intel VTune Amplifier

Paulius Velesko from Intel gave this talk at the ALCF Many-Core Developer Sessions. “This talk covers efficient profiling techniques that can help to dramatically improve the performance of code by identifying CPU and memory bottlenecks. Efficient profiling techniques can help dramatically improve the performance of code by identifying CPU and memory bottlenecks. We will demonstrate how to profile a Python application using Intel VTune Amplifier, a full-featured profiling tool.”

Jülich Supercomputers Power New Insights into Brain Imaging

Researchers are using biophysical modeling and simulations on Jülich supercomputers to develop new brain tissue imaging methods. “When generating a detailed network model of the brain, nerve fiber crossings pose a major challenge for current neuroimaging techniques. Scientists at Forschungszentrum Jülich have now found that scattered light can be used to resolve the brain’s substructure like the crossing angles of the nerve fibers with micrometer resolution. For their studies, the researchers combined microscopy measurements and simulations on supercomputers.”

Barcelona Supercomputing Centre to Optimize Storage and Data Analysis with PPI4HPC

The Barcelona Supercomputing Center (BSC) will provide a new storage infrastructure for enhanced data analysis capabilities thanks to the PPI4HPC (Public Procurement of Innovations for High Performance Computing). “The proposed infrastructure includes a disk tier built on all-flash technology and spinning disk drives. It changes the way underlying physical drives are managed, accessed and rebuilt in case of failures in order to minimize the impact to the scientific applications. It is a High-Performance Analytics compute infrastructure to run data analytics operations with the latest Power microprocessors and NVMe PCIe local storage to accelerate the workloads.”

NVIDIA Adds GPU and AI Expertise to COVID-19 HPC Consortium

A task force of NVIDIA computer scientists has joined the COVID-19 High Performance Computing Consortium, which brings together leaders from the U.S. government, industry and academia to accelerate research using the world’s most powerful HPC resources. “The consortium’s objective is to accelerate development of effective methods to detect, contain and treat the coronavirus. It will support researchers by providing access to 30 supercomputers with over 400 petaflops of compute performance.”

New Library of Artificial Antibodies Could Target Pathogens With Molecular Precision

A research team led by Berkeley Lab has developed a technique that could accelerate the design of artificial antibodies for biomedical applications – from sensing technologies that detect and neutralize infectious viruses and bacteria to the early detection of Alzheimer’s. “We can now readily build populations of rugged synthetic materials that can be engineered to recognize a potential pathogen,” said Zuckermann. “It is a shining example of biomimetic nanoscience.”

Video: High-Performance Memory For AI And HPC

In this video, Frank Ferro from Rambus examines the current performance bottlenecks in HPC, drilling down into power and performance for different memory options. “HBM2E offers the capability to achieve tremendous memory bandwidth. Four HBM2E stacks connected to a processor will deliver over 1.6 TB/s of bandwidth. And with 3D stacking of memory, high bandwidth and high capacity can be achieved in an exceptionally small footprint. Further, by keeping data rates relatively low, and the memory close to the processor, overall system power is kept low.”

Podcast: How crowd-sourced supercomputing is helping fight COVID-19

In this Roadhouse podcast, Dr. Greg Bowman from Folding @ Home describes the how crowdsourced computing is being used to fight the coronavirus and how we can get involved. “We are excited to announce a new batch of small molecule screening simulations are now up and running on Folding@home! These simulations will help prioritize which molecules will be synthesized and assayed by the COVID Moonshot aiming to rapidly developing new therapies against the SARS-CoV-2 main viral protease.”

NERSC Supercomputer to Help Fight Coronavirus

“NERSC is a member of the COVID-19 High Performance Computing Consortium. In support of the Consortium, NERSC has reserved a portion of its Director’s Discretionary Reserve time on Cori, a Cray XC40 supercomputer, to support COVID-19 research efforts. The GPU partition on Cori was installed to help prepare applications for the arrival of Perlmutter, NERSC’s next-generation system that is scheduled to begin arriving later this year and will rely on GPUs for much of its computational power.”