Introduction

Shoc Platform is a powerful, extensible system designed for running High Performance Computing (HPC) and Machine Learning (ML) workloads at scale. Whether you’re training large-scale AI models or running tightly coupled scientific simulations, Shoc provides the flexibility, performance, and control needed to orchestrate complex computational jobs efficiently.

What is Shoc Platform?

At its core, Shoc Platform is a job orchestration platform built for modern infrastructure. It simplifies the process of submitting, monitoring, and managing distributed compute jobs across dynamic clusters—whether they’re running on bare metal, virtual machines, or in the cloud.

Shoc Platform is tailored for teams and individuals who need to:

Run GPU-accelerated ML training and inference jobs
Launch distributed MPI/HPC workloads
Manage clusters and compute nodes programmatically
Maintain security and reproducibility across environments

Core Features

ML and HPC Job Orchestration

Define, submit, and monitor jobs using a unified CLI.
Support for containerized workloads with advanced resource scheduling.
Monitoring for CPU and memory usage.

Cluster Management

Register and manage compute clusters from multiple environments.
Dynamically allocate jobs to available nodes based on resource availability.
Node-level metrics: capacity, usage, health, and time-series monitoring.

Secrets Management

Store and inject secrets (e.g., API keys, DB passwords) securely into jobs.
Integrates with environment variable injection and volume mounts.
Supports encrypted storage and RBAC-based access.

Image and Registry Support

Pull images from public and private registries.
Authenticate against container registries per user or organization.

Multi-User and Role-Based Access Control

Namespace isolation for projects and teams.
Fine-grained access controls for jobs, clusters, secrets, and registries.

Tooling and Interfaces

CLI-first experience for power users and CI/CD integration.
Web UI for job tracking, cluster status, and user management.
REST API for integrating Shoc with your systems and automation scripts.