nomad

Nomad is HashiCorp's workload orchestrator for deploying and managing applications across on-premises and cloud infrastructure at scale. It ships as a single, self-contained binary that combines resource management and scheduling in one system, requiring no external services for storage or coordination. Workload types: Through pluggable task drivers, Nomad can run Docker and Podman containers, non-containerized executables and Java applications, and virtual machines via QEMU. This lets teams run microservices, batch jobs, and legacy applications side by side on the same infrastructure without needing to containerize everything. Operations and reliability: Nomad automatically handles application, node, and driver failures. It is distributed and resilient, using leader election and state replication to provide high availability. The scheduler is optimistically concurrent, which the project states increases throughput and reduces latency; Nomad has been proven to scale to clusters of 10K+ nodes in real-world production environments. Multi-region and multi-cloud: Federation is supported out of the box, enabling application deployments across multiple regions and clouds. Hardware acceleration: Built-in device plugins automatically detect and utilize resources from hardware devices such as GPUs, FPGAs, and TPUs, targeting machine learning and AI workloads. Ecosystem integration: Nomad integrates with Terraform for provisioning, Consul for service discovery, and Vault for secrets management. A commercial edition, Nomad Enterprise, is also available. Platforms and licensing: Nomad runs on Linux, Windows, and macOS. The source code is licensed under BUSL-1.1. Getting started: The repository links to official product documentation, CLI and API references, plugin docs, and tutorials, including a getting-started track for setting up a local non-production cluster. Terraform manifests for bringing up development Nomad clusters on public clouds are included in the terraform directory. Production deployments are covered by a reference architecture document with recommended practices. Product documentation is maintained in a separate web-unified-docs repository, and a public roadmap outlines major features expected in upcoming releases, with the caveat that items are subject to change. Contribution guidelines and developer documentation are located in the contributing directory, and community discussion takes place on the HashiCorp Discuss forum.