Deploy ArangoDB
ArangoDB supports multiple deployment modes to meet the exact needs of your project for resilience and performance
For installation instructions, please refer to the Installation chapter.
For production deployments, please also carefully check the ArangoDB Production Checklist.
Deployment Modes
ArangoDB can be deployed in different configurations, depending on your needs.
Single Instance
A Single Instance deployment is the most simple way to get started. Unlike other setups, which require some specific procedures, deploying a stand-alone instance is straightforward and can be started manually or by using the ArangoDB Starter tool.
Active Failover
Active Failover deployments use ArangoDB’s multi-node technology to provide high availability for smaller projects with fast asynchronous replication from the leading node to multiple replicas. If the leader fails, then a follower takes over seamlessly.
Cluster
Cluster deployments are designed for large scale operations and analytics, allowing you to scale elastically with your applications and data models. ArangoDB’s synchronously-replicating cluster technology runs on premises, on Kubernetes, and in the cloud on ArangoGraph - ArangoDB’s fully managed service.
Clustering ArangoDB not only delivers better performance and capacity improvements, but it also provides resilience through replication and automatic failover. You can deploy systems that dynamically scale up and down according to demand.
OneShard
ArangoDB Enterprise Edition ArangoGraph
OneShard deployments are cluster deployments but with the data of each database restricted to a single shard. This allows queries to run locally on a single DB-Server node for better performance and with transactional guarantees similar to a single server deployment. OneShard is primarily intended for multi-tenant use cases.
Datacenter-to-Datacenter
ArangoDB Enterprise Edition
For cluster deployments, ArangoDB supports Datacenter-to-Datacenter Replication (DC2DC). You can use it as an additional security feature to replicate your entire cluster off-site to another datacenter. The leading datacenter asynchronously replicates the data and configuration to the other datacenter for disaster recovery.
How to deploy
There are different ways to set up and operate ArangoDB.
You can start all the needed server processes manually, locally or on different machines, bare-metal or in Docker containers. This gives you the most control but you also need to manually deal with upgrades, outages, and so on.
You can use the ArangoDB Starter (the arangodb executable) to mostly automatically create and keep deployments running, either bare-metal or in Docker containers.
If you want to deploy in your Kubernetes cluster, you can use the ArangoDB Kubernetes Operator (
kube-arangodb).
The fastest way to get ArangoDB up and running is to run it in the cloud - the ArangoGraph Insights Platform offers a fully managed cloud service, available on AWS and GCP.
Manual Deployment
Single Instance:
Active Failover:
Cluster:
Deploying using the ArangoDB Starter
Setting up an ArangoDB cluster, for example, involves starting various nodes with different roles (Agents, DB-Servers, and Coordinators). The starter simplifies this process.
The Starter supports different deployment modes (single server, Active Failover,
cluster) and it can either use Docker containers or processes (using the
arangod executable).
Besides starting and maintaining ArangoDB deployments, the Starter also provides various commands to create TLS certificates and JWT token secrets to secure your ArangoDB deployments.
The ArangoDB Starter is an executable called arangodb and comes with all
current distributions of ArangoDB.
If you want a specific version, download the precompiled executable via the GitHub releases page .
Single Instance:
Active Failover:
Cluster:
Run in the cloud
- AWS and Azure
- ArangoGraph Insights Platform , fully managed, available on AWS and GCP