How does elasticsearch store its data [ multi node architecture ]

Elastic Stack Elasticsearch
1

Hi all,
I'm a newbie in elasticsearch, and i have a question about how does elasticsearch store its data in a multi node architecture, what i know is each node store its data on the specified data path and if we choose multiple data paths the data will be distributed between all paths, but what if i have a multi node architecture - each node is running on a server ( 200 gb disk ) and i have 8 nodes ( 8 servers ) is data going to be distributed and i'm going to have 200 * 8 gb of storage or it's going to be replicated and then i'll have just 200 gb.

Thank you very much in advance.

2

Welcome.

Let's take this scenario/

  • 8 nodes
  • 1 index with 8 primary shards and 0 replica shard
  • data size: 80gb

Each shard will have around 10gb of data (80/8).
You will have one shard allocated per node.
Each node will use then 10gb of disk space.

If you define the number of replicas to 1, you will have then in total 16 shards (8 primary shards + 8 replica shards).

So each node will hold one primary and one replica.
Each node will use 20gb of disk space.

HTH

2 Likes
3

Thank you very much for this detailed response :+1:

4

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.