https://github.com/deepsquare-io/grid

The DeepSquare Grid is a decentralized HPC based on Blockchain, in Solidity and Go, with an abstracted SLURM interface and meta-scheduling strategy.
https://github.com/deepsquare-io/grid

distributed go hpc supercomputers web3

Last synced: about 1 month ago
JSON representation

The DeepSquare Grid is a decentralized HPC based on Blockchain, in Solidity and Go, with an abstracted SLURM interface and meta-scheduling strategy.

• Host: GitHub
• URL: https://github.com/deepsquare-io/grid
• Owner: deepsquare-io
• Created: 2023-09-26T14:30:07.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-06-13T01:16:14.000Z (7 months ago)
• Last Synced: 2024-06-22T16:45:42.294Z (7 months ago)
• Topics: distributed, go, hpc, supercomputers, web3
• Language: Go
• Homepage: https://docs.deepsquare.run
• Size: 12.3 MB
• Stars: 10
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • Code of conduct: CODE_OF_CONDUCT.md
  • Codeowners: CODEOWNERS

Awesome Lists containing this project

README

        
# The DeepSquare Grid

## Overview

This repository is where [DeepSquare](https://deepsquare.io) develops the Grid. DeepSquare aggregates the unique capabilities of supercomputers from diverse compute providers worldwide into a unified computational infrastructure known as the [DeepSquare Grid](https://docs.deepsquare.run/workflow/learn/core-concepts#deepsquare-grid). Our primary aim is to democratize HPC, and the Grid stands as a powerful embodiment of this commitment.

DeepSquare operates with a user-centric approach to resource allocation. Users specify the computational requirements for their workloads, and DeepSquare, through its [Meta-scheduling](https://docs.deepsquare.run/workflow/learn/core-concepts#meta-scheduling) process, matches these workloads to the most appropriate compute provider available on the grid.

## Key features

With the rise of Machine Learning, DeepSquare is your gateway to the world of High-Performance Computing:

- **Global HPC Network**: DeepSquare Grid aggregates supercomputers from diverse global providers.

- **Democratized HPC**: DeepSquare Grid provides access to any users with specific computational requirements.

- **Tailored Resource Allocation:** DeepSquare Grid allows you to precise your workload's computational requirements, and let DeepSquare's Meta-Scheduling process find the perfect compute provider on the grid.

- **Effortless Efficiency:** Whether you're a scientist, researcher, or engineer, DeepSquare ensures your projects run smoothly and efficiently on cutting-edge HPC infrastructure.

- **Simplified Job Submission:** DeepSquare offers a straightforward API to launch various job types, including MPI and containers on the complex scheduler SLURM, abstracting the infrastructure and software complexity for many users.

## Getting started

If you'd like to try running a job on DeepSquare, you should start by reading our [Quick Start Guide](https://docs.deepsquare.run/workflow/getting-started/introduction) and our [documentation](https://docs.deepsquare.run)!

## Join the Community

- [Community Discord](https://discord.gg/KYWh28BkUE) - Request for support and help from the DeepSquare community.

- [Github Issues](https://github.com/deepsquare-io/grid/issues) - Submit your issues and feature requests via GitHub.

We welcome your help in building DeepSquare! If you are interested, we invite you to check

out the [Contributing Guide](./CONTRIBUTING.md).

## Documentation

Comprehensive details, tutorials, and reference materials are available at our [Documentation](https://docs.deepsquare.run/). Whether you're a new user getting started or an experienced developer looking for in-depth information, our documentation is the go-to resource to explore the full potential of DeepSquare.