Efficient Heterogeneous Parallel Programming Using OpenMP

This article is taken from Intel “Efficient Heterogeneous Parallel Programming Using OpenMP”. In this article, we will show you how to do CPU+GPU asynchronous calculations using OpenMP.

In some cases, offloading computations to an accelerator like a GPU means that the host CPU sits idle until the offloaded computations are finished. However, using the CPU and GPU resources simultaneously can improve the performance of an application. In OpenMP® programs that take advantage of heterogenous parallelism, the master clause can be used to exploit simultaneous CPU and GPU execution. In this article, we will show you how to do CPU+GPU asynchronous calculation using OpenMP.

The Intel® oneAPI DPC++/C++ Compiler was used with following command-line options:
‑O3 ‑Ofast ‑xCORE‑AVX512 ‑mprefer‑vector‑width=512 ‑ffast‑math ‑qopt‑multiple‑gather‑scatter‑by‑shuffles ‑fimf‑precision=low
‑fiopenmp ‑fopenmp‑targets=spir64=”‑fp‑model=precise”

OpenMP provides true asynchronous, heterogeneous execution on CPU+GPU systems. It’s clear from our timing results and VTune profiles that keeping the CPU and GPU busy in the OpenMP parallel region gives the best performance. We encourage you to try this approach.

Intel: Efficient Heterogeneous Parallel Programming Using OpenMP (Best Practices to Keep the CPU and GPU Working at the Same Time)


Compiling ORCA-4.2.1 with OpenMPI-3.1.4

ORCA is a general-purpose quantum chemistry package that is free of charge for academic users. The Project and Download Website can be found at ORCA Forum

You have to register yourself before you can participate in the forum or download ORCA-4.2.1. The current latest version for ORCA is 5.0.3. The package you might want to consider is ORCA 4.2.1, Linux, x86-64, .tar.xz Archive

Prerequisites that I use.

Unpacking ORCA-4.2.1

% tar -xvf orca_4_2_1_linux_x86-64_openmpi314.tar.xz

Running ORCA. If your environment has Module Environment

% module load openmpi/3.1.4/gcc-6.5.0

If not, you have to pacify PATH and LD_LIBRARY_PATH, MANPATH


Typical Input file

Calling ORCA requires full pathing

/usr/local/orca_4_2_1_linux_x86-64_openmpi314/orca $INPUT > $OUTPUT "--bind-to core --verbose"

For Input File usage, you may want to take a look at the ORCA 4.2.1 Manual found when you unpack or you can look at it online at orca_manual_4_2_1.pdf (enea.it) .

For example…….

! B3LYP def2-SVP SP
tda false
nroots 50
triplets true
nprocs 32

* xyz 0 1 fac_irppy3.xyz
  Ir        0.00000        0.00000        0.03016
   N       -1.05797        1.55546       -1.09121
   N        1.87606        0.13850       -1.09121

High-Severity Zero-Day Bug in Google Chrome

This article is taken from Singapore Computer Emergency Response Team (SINGCERT) titled High-Severity Zero-Day Bug in Google Chrome

Google has released Chrome 99.0.4844.84 for Windows, Mac, Linux and Chrome 99.0.4844.88 for Android users to address a high-severity zero-day bug (CVE-2022-1096)The vulnerability is a Type Confusion in V8 JavaScript engine exploit, and is reported to exist in the wild. V8 is Chrome’s component that is responsible for processing JavaScript code.

Type confusion refers to coding bugs during which an application initialises data execution operations using input of a specific “type” but is tricked into treating the input as a different “type”. This leads to logical errors in the application’s memory, which may allow an attacker to run unrestricted malicious codes inside an application.

No further technical details about the bug have been published by Google.

Google Chrome users on Windows, Mac and Linux are advised to upgrade to Chrome 99.0.4844.84 immediately by going into Chrome menu > Help > About Google Chrome, while Android users may refer to the Google Play Store for Chrome 99 (99.0.4844.88) version.

High-Severity Zero-Day Bug in Google Chrome

Compiling pybind11 with GNU-6.5

The Project Website can be found at https://github.com/pybind/pybind11

pybind11 is a lightweight header-only library that exposes C++ types in Python and vice versa, mainly to create Python bindings of existing C++ code. Its goals and syntax are similar to the excellent Boost.Python library by David Abrahams: to minimize boilerplate code in traditional extension modules by inferring type information using compile-time introspection.

The Compiling Steps can be found at https://pybind11.readthedocs.io/en/stable/basics.html

mkdir build
cd build
make check -j 4

Managing of Roaming Users’ Home Directories with Systemd-Homed

This article can be taken from OpenSource.com titled “Manage Linux users’ home directories with systemd-homed

Image By: OpenSource.com

The systemd-homed service supports user account portability independent of the underlying computer system. A practical example is to carry around your home directory on a USB thumb drive and plug it into any system which would automatically recognize and mount it. According to Lennart Poettering, lead developer of systemd, access to a user’s home directory should not be allowed to anyone unless the user is logged in. The systemd-homed service is designed to enhance security, especially for mobile devices such as laptops. It also seems like a tool that might be useful with containers.

This objective can only be achieved if the home directory contains all user metadata. The ~/.identity file stores user account information, which is only accessible to systemd-homed when the password is entered. This file holds all of the account metadata, including everything Linux needs to know about you, so that the home directory is portable to any Linux host that uses systemd-homed. This approach prevents having an account with a stored password on every system you might need to use.

The home directory can also be encrypted using your password. Under systemd-homed, your home directory stores your password with all of your user metadata. Your encrypted password is not stored anywhere else thus cannot be accessed by anyone. Although the methods used to encrypt and store passwords for modern Linux systems are considered to be unbreakable, the best safeguard is to prevent them from being accessed in the first place. Assumptions about the invulnerability of their security have led many to ruin.

This service is primarily intended for use with portable devices such as laptops. Poettering states, “Homed is intended primarily for client machines, i.e., laptops and thus machines you typically ssh from a lot more than ssh to, if you follow what I mean.” It is not intended for use on servers or workstations that are tethered to a single location by cables or locked into a server room.

The systemd-homed service is enabled by default on new installations—at least for Fedora, which is the distro that I use. This configuration is by design, and I don’t expect that to change. User accounts are not affected or altered in any way on systems with existing filesystems, upgrades or reinstallations that keep the existing partitions, and logical volumes.

Manage Linux users’ home directories with systemd-homed (OpenSource.com)

For more Read-Up, do take a look at “Manage Linux users’ home directories with systemd-homed

Licensing Errors for ANSYS IcePak Solver

If you are encountering issues like the one below.

"If you have this error "Failed to enable features using current license settings. Note that Pro, Premium, Enterprise licenses are available on your server. To use these licenses check the corresponding option. For more information, search "PPE" in the help documentation. Failover feature "ANSYS IcePak Solver" is not available. Request name aice_solv does not exist in the....."

Highlighted Area of Research by HPCWire

HPCWire highlighted 3 Areas of Research in the High-Performance Computing and Related Domains. The article can be found here

Research 1: HipBone: A performance-portable GPU-accelerated C++ version of the NekBone benchmark

HipBone “is a fully GPU-accelerated C++ implementation of the original NekBone CPU proxy application with several novel algorithmic and implementation improvements which optimize its performance on modern finegrain parallel GPU accelerators.” 

What’s New in HPC Research: HipBone, GPU-Aware Asynchronous Tasks, Autotuning & More

Research 2: A Case for intra-rack resource disaggregation in HPC

A multi-institution research team utilized Cori, a high performance computing system at the National Energy Research Scientific Computing Center, to analyze “resource disaggregation to enable finer-grain allocation of hardware resources to applications.”

What’s New in HPC Research: HipBone, GPU-Aware Asynchronous Tasks, Autotuning & More

Research 3: Improving Scalability with GPU-Aware Asynchronous Tasks

Computer scientists from the University of Illinois at Urbana-Champaign and Lawrence Livermore National Laboratory demonstrated improved scalability to hide communication behind computation with GPU-aware asynchronous tasks.

What’s New in HPC Research: HipBone, GPU-Aware Asynchronous Tasks, Autotuning & More