PackageManager_EN - CCSEPBVR/CS-IS-PBVR GitHub Wiki

Package Manager

Introduction

CS-PBVR supports the package managers apt, yum, Homebrew and Spack.
This section explains how to install CS-PBVR using apt, yum, Homebrew and Spack.

Installing with apt/yum(CS-PBVR)

CS-PBVR can be installed using the apt/yum package management systems available on Linux.
apt is available on Debian, Ubuntu, and similar distributions.
yum (dnf) is available on Red Hat Enterprise Linux, CentOS, and similar distributions.
The packages were built with glibc 2.28. They cannot run on versions older than glibc 2.28.
You can check your glibc version with the following command:

ldd --version

The apt package was built on Ubuntu 18.10.
The yum package was built on CentOS 8.5.
Both packages were built for the x86_64 (amd64) architecture.

Install and Uninstall with apt

Create a new file at /etc/apt/sources.list.d/pbvr.list, paste the following content into it, and save it:

deb [arch=amd64 trusted=yes] https://ccsepbvr.github.io/apt-yum-pbvr/apt stable main

Update the package list with the following command:

sudo apt update

Then install the package with:

sudo apt install -y pbvr

Uninstall the package using:

sudo apt purge pbvr
sudo apt autoremove  # Includes dependent packages

Remove the added repository with:

sudo rm /etc/apt/sources.list.d/pbvr.list
sudo apt update

Install and Uninstall with yum (dnf)

Create a new file at /etc/yum.repos.d/pbvr.repo, paste the following content into it, and save it:

[pbvr-repo]
name=pbvr
baseurl=https://ccsepbvr.github.io/apt-yum-pbvr/yum
enabled=1
gpgcheck=0

Update the package cache with:

sudo dnf makecache

Then install the package using either of the following commands:

sudo yum install -y pbvr
sudo dnf install -y pbvr

Uninstall the package using:

sudo dnf remove pbvr
sudo dnf autoremove  # Includes dependent packages

Remove the added repository with:

sudo rm /etc/yum.repos.d/pbvr.repo
sudo dnf clean all
sudo dnf makecache

Execution

The following commands become available.
Usage of the server, client, and filter program is described here.
Usage of the KVSML converter is described here.
Sample data is available here.

pbvr_server # Server program
pbvr_client # Client program
pbvr_filter # Filter program
kvsml-converter # KVSML converter

Installing with Homebrew (CS-PBVR)

You can install CS-PBVR using Homebrew, a package management tool available on macOS and Linux.
The package is created using Homebrew’s Tap feature and is not an official package.
Homebrew has its own requirements.
It also defines Support Tiers, and Tier 1 is supported.

Installation

Run the following commands to install. You can choose either the standard version or the extended file format version.
The extended file format version allows you to visualize sample data without using the filter program or KVSML converter.
For large-scale data, it is recommended to use the filter program and KVSML converter to convert the data into KVSML format for visualization.

brew tap ccsepbvr/pbvr
brew install pbvr # Normal version
brew install pbvr-extended-fileformat # Extended file format version

Execution

The following commands become available.
Usage of the server, client, and filter program is described here.
Usage of the KVSML converter is described here.
Sample data is available here.

pbvr_server # Server program
pbvr_client # Client program
pbvr_filter # Filter program
kvsml-converter # KVSML converter

Installing with Spack (CS-PBVR)

You can install CS-PBVR using Spack, a package management tool for HPC environments.
Spack allows adding repositories, and by adding an original repository, CS-PBVR can be installed.

Official Spack and Fujitsu Compiler-Compatible Spack

There are two versions of Spack: the official one, and one adapted for the Fujitsu compiler (used on systems such as Fugaku).
The Fujitsu compiler-compatible version of Spack is developed by RIKEN, forked from the official Spack.
The official Spack repository on GitHub is here.
The Fujitsu compiler-compatible Spack repository is here.
The original repository to be added differs between the official Spack and the Fujitsu compiler-compatible Spack.
Original repository for official Spack is here.
Original repository for Fujitsu compiler-compatible Spack is here.

Verified Compilers

Installation has been confirmed with the following compilers:

Verified Supercomputers

Installation has been confirmed on the following supercomputers:

  • SGI8600 (JAEA, official Spack)
  • Wisteria (University of Tokyo, Fujitsu compiler-compatible Spack)
  • Miyabi (University of Tokyo, official Spack)
  • Fugaku (RIKEN, Fujitsu compiler-compatible Spack)

Installation

To install using official Spack

git clone https://github.com/CCSEPBVR/spack-pbvr.git # Clone repository
spack repo add /path/to/spack-pbvr # Add repository to Spack
spack install pbvr %[email protected] # Install with [email protected] compiler

To install using Fujitsu compiler-compatible Spack

git clone https://github.com/CCSEPBVR/spack-pbvr-fujitsu.git # Clone repository
spack repo add /path/to/spack-pbvr-fujitsu # Add repository to Spack
spack install pbvr %[email protected] # Install with [email protected] compiler
spack install pbvr %[email protected] # Install with [email protected] compiler

Installation Options

Options can be set during installation.
All options are enabled by default.
The Fujitsu compiler ([email protected]) currently does not support client and extended_fileformat.

  • client: Build the client
  • mpi: Enable MPI parallelization
  • extended_fileformat: Enable extended file formats (VTK)

Example of setting options:

spack install pbvr +client +mpi +extended_fileformat %[email protected] # Enable all (default)
spack install pbvr ~client ~mpi ~extended_fileformat %[email protected] # Disable all

When the extended file format (VTK) is enabled, sample data can be visualized without using the filter program or KVSML converter.
For large-scale data, it is recommended to use the filter program and KVSML converter to convert the data into KVSML format for visualization.

Execution

The following commands become available.
Usage of the server, client, and filter program is described here.
Usage of the KVSML converter is described here.
Sample data is available here.

pbvr_server # Server program
pbvr_client # Client program
pbvr_filter # Filter program
kvsml-converter # KVSML converter
⚠️ **GitHub.com Fallback** ⚠️