# Replication with Memgraph Community Replication allows you to **copy data from one Memgraph instance (MAIN) to one or more REPLICA instances**. This improves **read scalability**, provides **extra copies of your data** and can reduce the impact of individual node failures. In a replication setup: - The **MAIN** instance accepts **reads and writes**. - **REPLICA** instances accept **reads only** and continuously replicate data from the MAIN. Replication is a good fit if you: - Want to scale read workloads horizontally. - Need additional copies of your data for resilience. - Are comfortable managing **manual failover** and cluster orchestration yourself. > **Info** > > This guide is for **Memgraph Community** users who want to set up data > replication across multiple instances. If you have a **Memgraph Enterprise** > license, we strongly recommend using the [high availability > features](https://memgraph.com/docs/clustering/high-availability) instead. > > High availability (HA) builds on replication and adds: > - **Automatic failover** (no manual promotion of replicas), > - **Coordinator instances** based on the Raft consensus protocol, > - **Built-in routing** and better protection against data loss, > - A more robust platform for true **24/7 uptime**. > **Note** > > Before you start, we recommend reading **[replication under the hood > section](https://memgraph.com/docs/clustering/replication/how-replication-works)** to understand the > underlying concepts: roles (MAIN / REPLICA), replication modes (**SYNC**, > **ASYNC**, **STRICT_SYNC**), durability (snapshots, WAL, deltas), and replica > recovery behavior. --- ## What you'll find in this section This **Replication** section focuses on the *practical* side of setting up and operating a replication cluster. Use it as your entry point to: - Choose a deployment method (Docker or Kubernetes), - Apply best practices for reliability and performance, - Use the available commands to manage and inspect the cluster. ### [How replication works](https://memgraph.com/docs/clustering/replication/how-replication-works) Learn how Memgraph replication works under the hood, including: - CAP theorem trade-offs - Roles of **MAIN** and **REPLICA** - How data is replicated using deltas, WALs, and snapshots - Replica states and recovery - Replication modes (**ASYNC**, **SYNC**, **STRICT_SYNC**) and their guarantees - How replicas catch up after interruptions - Multi-tenant replication (Enterprise) - Key internals such as timestamps, epoch IDs, and durability handling A deeper look at how Memgraph maintains consistency, performance, and resilience across replicated instances. ### [Set up replication cluster using Docker](https://memgraph.com/docs/clustering/replication/setup-replication-cluster-docker) Create a local replication cluster using Docker. This guide walks through running one MAIN and two REPLICA instances, configuring SYNC/ASYNC modes, and verifying that data is replicated correctly. This is the recommended starting point if you are experimenting locally or running simple container-based deployments. ### [Set up replication cluster on K8s](https://memgraph.com/docs/clustering/replication/setup-replication-cluster-k8s) Deploy a replication cluster on Kubernetes using your own manifests or the standalone Helm chart. Memgraph currently **does not** provide official Helm charts for the Community edition, so this guide focuses on using raw Kubernetes manifests and concepts. ### [Replication best practices](https://memgraph.com/docs/clustering/replication/best-practices) Guidelines for choosing replication modes, configuring storage, sizing hardware, setting flags, managing manual failover, and running replication safely in Memgraph Community. ### [Replication commands reference guide](https://memgraph.com/docs/clustering/replication/replication-commands-reference) A complete list of all Community Edition replication commands, including role changes, replica registration, monitoring, and cluster maintenance operations.