Nomad (HashiCorp)

Nomad (HashiCorp)
Name	Nomad
Developer	HashiCorp
Released	2015
Programming language	Go
Operating system	Linux, Windows, macOS
License	Mozilla Public License 2.0
Website	HashiCorp product pages

Nomad (HashiCorp) Nomad is a workload orchestration and cluster manager created by HashiCorp and released in 2015. It schedules containerized and non-containerized applications across heterogeneous infrastructure, integrating with projects and organizations such as Docker, Kubernetes, Consul, Vault, and cloud providers including Amazon Web Services, Microsoft Azure, and Google Cloud Platform. Nomad is designed for simplicity, scalability, and multi-datacenter operation, and is used by enterprises, research institutions, and government agencies that run distributed systems such as GitLab, Shopify, Adobe, and Ticketmaster.

Overview

Nomad is a cluster scheduler that supports multiple workload types including containers, virtual machines, and standalone binaries. It was developed by HashiCorp, the company behind Vagrant, Terraform, Consul, and Vault. Nomad competes and interoperates with orchestration projects like Kubernetes, Mesos, and Docker Swarm. Its design emphasizes a single binary written in Go, an agent architecture influenced by distributed systems research such as the Raft consensus algorithm and projects from institutions like MIT, Berkeley, and Stanford University.

Architecture

Nomad uses a client–server model with servers forming a consensus-based control plane and clients running workloads. The server nodes use consensus protocols inspired by Raft to provide leader election and state replication comparable to systems used by etcd and Consul. Clients interact with server leaders and integrate with service discovery systems such as Consul and secret managers like Vault. Nomad jobs are defined in a declarative HCL format familiar to users of Terraform, and the scheduler supports binpacking, spread, and affinity strategies akin to strategies in Apache Mesos and Kubernetes. Nomad also supports federation across multiple datacenters, enabling deployments that span providers like Amazon Web Services, Microsoft Azure, Google Cloud Platform, and on-premises environments such as VMware ESXi clusters.

Features

Nomad provides multi-region and multi-datacenter scheduling, workload diversification, and autoscaling integrations. It supports task drivers for Docker, rkt, qemu, JVM processes, and raw executables, similar to heterogeneous workload support found in Apache Mesos. Nomad offers job grouping, periodic jobs, and system jobs for cluster services analogous to features in Kubernetes CronJobs and DaemonSets. Native integrations with Consul enable service registration, while integration with Vault and identity providers such as Okta and Azure Active Directory facilitate secret injection and authentication. Observability is supported via telemetry hooks compatible with Prometheus, Grafana, Datadog, and logging systems used by Splunk and ELK Stack projects.

Use cases and adoption

Organizations use Nomad for continuous delivery pipelines, batch processing, and hybrid cloud migration. Companies like HashiCorp, Monzo, and Electronic Arts have published case examples using Nomad to run CI/CD workloads, game servers, and data pipelines. Nomad is adopted in sectors including fintech, e-commerce, media, and government, where interoperability with tools from HashiCorp’s ecosystem or cloud providers such as Amazon Web Services is important. It is chosen where teams require simple operational models compared with the complexity of Kubernetes or where organizations need to schedule legacy applications alongside modern microservices, similar to use cases described by adopters of Apache Mesos and Docker Swarm.

Comparison with other orchestrators

Compared with Kubernetes, Nomad offers a smaller operational surface and a single binary architecture, trading off some ecosystem size for lower complexity and faster upgrades. Relative to Apache Mesos, Nomad provides more opinionated scheduling primitives for modern DevOps workflows. Against Docker Swarm, Nomad offers multi-region capabilities and richer workload types beyond containers. Nomad interoperates with Consul for service discovery in a way comparable to integrations between Kubernetes and etcd, and Nomad’s integration story parallels how Terraform integrates with cloud providers like Amazon Web Services, Microsoft Azure, and Google Cloud Platform.

Deployment and operation

Nomad clusters are typically deployed across availability zones and regions; operational patterns borrow from enterprise practices used by Netflix, Google, and Facebook for high availability. Production deployments often combine Nomad with Consul for discovery and Vault for secrets, while observability uses Prometheus exporters and dashboards in Grafana. Teams use infrastructure-as-code tools such as Terraform to provision compute resources on Amazon Web Services, Microsoft Azure, Google Cloud Platform, and VMware ESXi before installing Nomad. Upgrade strategies and rolling restarts follow patterns established by large-scale services like Kubernetes control plane upgrades and use leader election semantics like those in Raft.

Security and compliance

Nomad supports mTLS for client–server and server–server communication, enabling encryption comparable to transports in Consul and etcd. Access control is provided via ACLs and integration with identity providers such as Okta and Azure Active Directory; secrets management leverages Vault for dynamic credential issuance, mirroring practices in secure deployments used by organizations like Capital One and GitHub. Compliance-oriented deployments map Nomad auditing and telemetry to systems used in regulated industries, aligning with frameworks and standards adopted by NIST and security practices advocated by vendors such as Red Hat and IBM.

Category:HashiCorp