DeepSquare offers a decentralized high‑performance computing platform that lets users submit jobs via a simple YAML workflow, abstracting complex runtimes and scripts. The ecosystem leverages blockchain‑based scheduling to provide transparent, fair access to GPU, CPU and accelerator resources across a distributed network. Compute resources are paid for and rewarded with the DPS token on the Avalanche network, enabling sustainable and professional HPC for AI, rendering, and simulation workloads.
Funding
$2.5M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Accessing high-performance computing (HPC) resources often requires navigating complex infrastructure and software configurations, limiting accessibility for many users. Traditional centralized cloud providers can be expensive and may lack transparency in resource allocation.
Solution
DeepSquare provides a decentralized HPC ecosystem that aggregates computing power from various providers into a unified grid, making HPC resources more accessible. The platform simplifies job submission by abstracting runtimes into a user-friendly YAML format, replacing complex shell scripts. DeepSquare's meta-scheduler, deployed on a blockchain, ensures transparent and equitable resource allocation by matching user workloads to the most suitable compute provider. The platform supports containerization, parallelization technologies (MPI, NCCL), and hardware accelerators, enabling users to run a wide range of computing needs, including AI, rendering, and simulations.
Target Audience
The primary target audience includes AI engineers, researchers, developers, and businesses requiring high-performance computing resources for AI, rendering, simulations, and other computationally intensive tasks.
Features
- Decentralized HPC grid aggregating supercomputers from diverse providers
- Simplified job submission using a YAML-based workflow definition
- Transparent and fair resource allocation via a blockchain-based meta-scheduler
- Support for containerization (OCI containers) and parallelization technologies (MPI, NCCL)
- Compatibility with hardware accelerators, including GPUs and Infiniband
- Support for interactive workflows with external network access
- Billed per actual usage of compute resources, down to the minute
- Kubernetes-based infrastructure orchestrator automates the deployment, management, and monitoring of compute-optimized clusters