Hedgehog

About Hedgehog

Hedgehog provides open source software that enables Cloud Native application owners to deploy workloads on edge compute and distributed cloud infrastructure with high effective bandwidth and low latency, optimizing AI training and inference. The platform simplifies network operations by automating congestion management and routing in GPU fabrics, eliminating the need for specialized network engineers.

```xml <problem> AI training and inference workloads require high bandwidth and low latency, but traditional data center networks often suffer from congestion and packet loss, leading to increased job completion times and a poor user experience. Operating these networks is complex, frequently requiring specialized network engineers and expensive proprietary hardware. </problem> <solution> Hedgehog provides open-source software that enables cloud-native application owners to deploy AI workloads on edge compute and distributed cloud infrastructure with optimized network performance. The platform automates congestion management and routing within GPU fabrics, increasing effective bandwidth and reducing latency. By simplifying network operations, Hedgehog eliminates the need for specialized network engineers, offering a cloud-like user experience for managing AI infrastructure. </solution> <features> - Automated congestion management and adaptive routing for optimal AI network performance using RoCEv2, ECN, and PFC. - Open-source software allowing for hardware vendor flexibility and reduced costs. - Cloud-like user experience familiar to existing cloud operations teams, eliminating the need for specialized network engineers. - Virtual private cloud (VPC) service for north-south traffic to GPU clusters, providing multi-tenant cloud services similar to AWS, Azure, or Google Cloud Platform. - Gateway services for internet connectivity and integration with other cloud service providers, including transit, load balancing, and security features. - Integration with SONiC (Software for Open Networking in the Cloud). </features> <target_audience> The primary target audience includes AI cloud builders and cloud-native application owners who need to deploy AI workloads on edge compute and distributed cloud infrastructure. </target_audience> ```

What does Hedgehog do?

Hedgehog provides open source software that enables Cloud Native application owners to deploy workloads on edge compute and distributed cloud infrastructure with high effective bandwidth and low latency, optimizing AI training and inference. The platform simplifies network operations by automating congestion management and routing in GPU fabrics, eliminating the need for specialized network engineers.

Where is Hedgehog located?

Hedgehog is based in Seattle, United States.

When was Hedgehog founded?

Hedgehog was founded in 2022.

How much funding has Hedgehog raised?

Hedgehog has raised 7380000.

Location
Seattle, United States
Founded
2022
Funding
7380000
Employees
19 employees
Major Investors
Engineering Capital
Looking for specific startups?
Try our free semantic startup search

Hedgehog

Score: 100/100
AI-Generated Company Overview (experimental) – could contain errors

Executive Summary

Hedgehog provides open source software that enables Cloud Native application owners to deploy workloads on edge compute and distributed cloud infrastructure with high effective bandwidth and low latency, optimizing AI training and inference. The platform simplifies network operations by automating congestion management and routing in GPU fabrics, eliminating the need for specialized network engineers.

githedgehog.com1K+
cb
Crunchbase
Founded 2022 – Seattle, United States

Funding

$

Estimated Funding

$7.4M+

Major Investors

Engineering Capital

Team (15+)

Steven Noble

Staff

Marc Austin

GPU Connective Tissue

Amit Limaye

staff

Company Description

Problem

AI training and inference workloads require high bandwidth and low latency, but traditional data center networks often suffer from congestion and packet loss, leading to increased job completion times and a poor user experience. Operating these networks is complex, frequently requiring specialized network engineers and expensive proprietary hardware.

Solution

Hedgehog provides open-source software that enables cloud-native application owners to deploy AI workloads on edge compute and distributed cloud infrastructure with optimized network performance. The platform automates congestion management and routing within GPU fabrics, increasing effective bandwidth and reducing latency. By simplifying network operations, Hedgehog eliminates the need for specialized network engineers, offering a cloud-like user experience for managing AI infrastructure.

Features

Automated congestion management and adaptive routing for optimal AI network performance using RoCEv2, ECN, and PFC.

Open-source software allowing for hardware vendor flexibility and reduced costs.

Cloud-like user experience familiar to existing cloud operations teams, eliminating the need for specialized network engineers.

Virtual private cloud (VPC) service for north-south traffic to GPU clusters, providing multi-tenant cloud services similar to AWS, Azure, or Google Cloud Platform.

Gateway services for internet connectivity and integration with other cloud service providers, including transit, load balancing, and security features.

Integration with SONiC (Software for Open Networking in the Cloud).

Target Audience

The primary target audience includes AI cloud builders and cloud-native application owners who need to deploy AI workloads on edge compute and distributed cloud infrastructure.