Rate This Document
Findability
Accuracy
Completeness
Readability

Introduction

etcd Overview

etcd is a highly consistent and distributed key-value storage service that provides a reliable way to store data that needs to be accessed in a distributed system or host cluster.

An etcd cluster usually consists of three or five nodes. A consensus algorithm Raft is used to implement consistency and collaboration of multiple nodes in a distributed system. The algorithm selects a leader node, and the other nodes are followers. The leader synchronizes and distributes data. It synchronizes logs to followers to ensure that each node in this cluster has a copy of data. The leader periodically sends heartbeat packets to all nodes. If a node does not receive heartbeat packets within a specified period, a new leader will be selected.

This document describes how to use a benchmark tool to test the read and write status of the deployed etcd cluster.

Recommended Version

etcd 3.1.20

Deploy and start an etcd cluster. For details, see Deploying an etcd Cluster in the etcd Cluster Deployment Guide.