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

Putting Computer Vision to Work with OpenVINO

OpenVINO is a single toolkit, optimized for Intel hardware, that the data scientist and AI software developer can use for quickly developing high-performance applications that employ neural network inference and deep learning to emulate human vision over various platforms. “This toolkit supports heterogeneous execution across CPUs and computer vision accelerators including GPUs, Intel® Movidius™ hardware, and FPGAs.”

NVIDIA steps up with Nsight Systems Performance Analysis Tool

Today NVIDIA announced that NVIDIA Nsight Systems 2019.1 is now available for download. As a system-wide performance analysis tool. With it, developers can visualize application algorithms, identify large optimization opportunities, and tune/scale efficiently across CPUs and GPUs. “In this release, we introduce a wide range of new features, refinements, and fixes. The enhancements aim to improve a user’s ability to analyze neural network performance, locate graphical stutter, and increase pattern discoverability.”

Video: Speeding up Programs with OpenACC in GCC

Thomas Schwinge from Mentor gave this talk at FOSDEM’19. “Requiring only few changes to your existing source code, OpenACC allows for easy parallelization and code offloading to accelerators such as GPUs. We will present a short introduction of GCC and OpenACC, implementation status, examples, and performance results.”

Argonne Looks to Singularity for HPC Code Portability

Over at Argonne, Nils Heinonen writes that Researchers are using the open source Singularity framework as a kind of Rosetta Stone for running supercomputing code almost anywhere. “Once a containerized workflow is defined, its image can be snapshotted, archived, and preserved for future use. The snapshot itself represents a boon for scientific provenance by detailing the exact conditions under which given data were generated: in theory, by providing the machine, the software stack, and the parameters, one’s work can be completely reproduced.”

Video: OpenHPC Update

Adrian Reber from Red Hat gave this talk at the FOSDEM’19 conference. “In this talk I want to give an introduction about the OpenHPC project. Why do we need something like OpenHPC? What are the goals of OpenHPC? Who is involved in OpenHPC and how is the project organized? What is the actual result of the OpenHPC project? It also has been some time (it was FOSDEM 2016) since OpenHPC was part of the HPC, Big Data and Data Science devroom, so that it seems a good opportunity for an OpenHPC status update and what has happened in the last three years.”

Rapids: Data Science on GPUs

Christoph Angerer from NVIDIA gave this talk at FOSDEM’19. “The next big step in data science will combine the ease of use of common Python APIs, but with the power and scalability of GPU compute. The RAPIDS project is the first step in giving data scientists the ability to use familiar APIs and abstractions while taking advantage of the same technology that enables dramatic increases in speed in deep learning. This session highlights the progress that has been made on RAPIDS, discusses how you can get up and running doing data science on the GPU, and provides some use cases involving graph analytics as motivation.”

Are Platform Configuration Problems Degrading Your Application’s Performance?

The Intel VTune™ Amplifier Platform Profiler on Windows* and Linux* systems shows you critical data about the running platform that help identify common system configuration errors that may be causing performance issues and bottlenecks. Fixing these issues, or modifying the application to work around them, can greatly improve overall performance.

Video: EasyBuild State of the Union

In this video from the 2019 EasyBuild User Meeting, Kenneth Hoste from the University of Ghent presents: EasyBuild State of the Union. “EasyBuild is a software build and installation framework that allows you to manage (scientific) software on HPC systems in an efficient way. The EasyBuild User Meeting includes presentations by both EasyBuild users and developers, next to hands-on sessions.”

Accelerated Python for Data Science

The Intel Distribution for Python takes advantage of the Intel® Advanced Vector Extensions (Intel® AVX) and multiple cores in the latest Intel architectures. By utilizing the highly optimized Intel MKL BLAS and LAPACK routines, key functions run up to 200 times faster on servers and 10 times faster on desktop systems. This means that existing Python applications will perform significantly better merely by switching to the Intel distribution.

Call For Proposals: Worldwide GPU Hackathons in 2019

ORNL has issued its Call for Proposals for a set of global GPU Hackathons in 2019. “A GPU hackathon is a 5-day coding event in which teams of developers port their applications to run on GPUs, or optimize their applications that already run on GPUs. Each team consists of three or more developers who are intimately familiar with (some part of) their application, and they work alongside two mentors with GPU programming expertise. The mentors come from universities, national laboratories, supercomputing centers, government institutions, and vendors.”