Data

Tools

Indexes

Applications

Experiments

Open Source Science

qrackmin

ThereminQ Qrackmin QuantOPS : HPC Qrack container images

https://github.com/twobombs/qrackmin

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (7.3%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

ThereminQ Qrackmin QuantOPS : HPC Qrack container images

Basic Info
  • Host: GitHub
  • Owner: twobombs
  • License: gpl-3.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 1.19 MB
Statistics
  • Stars: 1
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 4
Created over 3 years ago · Last pushed 8 months ago
Metadata Files
Readme License Citation

README.md

Qrackmin - a minimalistic container for Qrack - OpenCL & CUDA


Screenshot from 2024-10-05 15-58-16 Qrackmin container system deployed in rancher through the ThereminQ HELM definitions

Docker deployments

  • The :latest container image is meant to be used on a single node with Nvidia-Docker2 and Linux support

bash docker run --gpus all --device=/dev/kfd --device=/dev/dri:/dev/dri --privileged -d twobombs/qrackmin[:tag] [--memory 24G --memory-swap 250G] `

  • -v /var/log/qrack:/var/log/qrack for saving of measured results outside container

bash docker exec -ti [containerID] bash `

  • ThereminQ repo with runfiles is checked out on /root

  • Windows users should install WSL2, Docker Desktop, docker.io, nvidia-docker2 to run this (CUDA only)

  • Qrackmin:AWS - Qrackmin:BRAKET

    on demand AWS template proposals for x86 and ARM - CUDA, OpenCL and CPU powered

  • :AWS boilerplate binary runtime code for Qrack as a Service - QFT RND benchmarks output

  • :BRAKET boilerplate python runtime code for PyQrack as a |BraKET> container service

  • ### Qrackmin:pyqrack - a python runtime environment to run tests for pyqrack
  • ### Qrackmin:qbdd - a python runtime environment to run benchmarks for qbdd
  • ### Qrackmin:elidded - Elidded and patched quadrant time tests

1734875657236

bash docker run --gpus all --device=/dev/dri:/dev/dri -d twobombs/qrackmin:elidded `

  • ### Qrackmin:POCL - latest & POCL added for CPU-only support

:pocl container image adds the generic OpenCL-ICD and is to be used with high memory & CPU count hosts

  • Simulate performance and measured results on CPU
  • For validation before GPU cluster deployment
  • ### Qrackmin:VCL - Qrack@VCL with node support

The :vcl tag contains VCL binaries that are copyrighted by Amnon Barak to run VCL as a backend and host

On the host the following directory structure needs to be created

sudo mkdir /var/log/vcl /var/log/vcl/etc /var/log/vcl/etc/vcl/ /var/log/vcl/etc/init.d /var/log/vcl/usr /var/log/vcl/usr/bin /var/log/vcl/etc/rc0.d /var/log/vcl/etc/rc1.d /var/log/vcl/etc/rc2.d /var/log/vcl/etc/rc3.d /var/log/vcl/etc/rc4.d /var/log/vcl/etc/rc5.d /var/log/vcl/etc/rc6.d

Run the bash in /run-scripts/ in this repository called ./1-run-nodes

You will be asked two questions: - the amount of virtual nodes you want to create - the NVIDIA devices you want to expose ( often 'all' will suffice, otherwise use the device number )*

When you've deployed enough backend containers to you liking you can start ./2-run-host

  • the nodes' IPs will be scraped
  • the host container will be started and will initialize
  • you'll drop into the host-containers' bash
  • then run workloads through ./vcl-1.25/vclrun [command]

(*) = other OpenCL device types such as an IntelIGP that are also recognised will also be taken into the cluster

A Full VDI host experience is avaliable in ThereminQ:vcl-controller

Follow the guide for running VCL in a cluster here

For multi-host setup please look Docker Swarm and Zerotier

Credits go to Dan Strano for creating Qrack and Qimcifa

Some rights are reserved regarding code and functionality for the Amazon AWS |BraKET> container images; they are in the git repo checkouts for the :aws and :braket container images and the aws tools for Amazon AWS as well

Owner

  • Name: Aryan Blaauw
  • Login: twobombs
  • Kind: user
  • Location: Hengelo (NL)

🧭 Teleporting 🌊 Hilbert Space 🌌 QuantOPS ⚒️ Bladerunner 🤖

Citation (CITATION.cff) 

# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
title: ThereminQ
message: >-
  If you use this software, please cite it using the
  metadata from this file.
type: software
authors:
  - given-names: Aryan Jacques
    family-names: Blaauw
    email: ablaauw@gmx.net
    orcid: 'https://orcid.org/0000-0002-5657-7824'
identifiers:
  - type: url
    value: 'https://arxiv.org/abs/2304.14969'
repository-code: 'https://github.com/twobombs/thereminq'
url: 'https://github.com/twobombs'
repository-artifact: 'https://hub.docker.com/r/twobombs/thereminq'
abstract: >-
  ThereminQ orchestrates a suite of best-of-class tools
  designed to control extend and visualize data emanating
  for Quantum circuits using Qrack ELK Tipsy Jupyter CUDA
  and OpenCL accelerators
keywords:
  - docker
  - cuda
  - quantum computing
  - opencl
license: GPL-3.0+
version: 4.x
date-released: '2023-05-23'

GitHub Events

Total
  • Push event: 155
Last Year
  • Push event: 155

Dependencies

dockerfiles/Dockerfile docker
  • nvidia/cudagl 11.4.2-base-ubuntu20.04 build