Index settings (shards, replicas) in Elasticsearch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a shard in Elasticsearch?

A shard is a basic unit of storage in Elasticsearch. It holds a subset of the data in an index, allowing the data to be split across multiple nodes for better performance and scalability.

beginner

What does the number_of_shards setting control?

It controls how many primary shards an index is divided into. This number cannot be changed after the index is created.

beginner

What is a replica in Elasticsearch?

A replica is a copy of a primary shard. It provides fault tolerance and increases search capacity by allowing queries to run on multiple copies.

intermediate

How does the number_of_replicas setting affect an index?

It sets how many copies of each primary shard are created. More replicas mean better fault tolerance and faster searches but use more storage.

intermediate

Can you change the number_of_replicas after creating an index?

Yes, you can change the number_of_replicas at any time to add or remove replica shards without downtime.

What happens if you increase the number_of_replicas for an index?

A More copies of shards are created, improving fault tolerance and search speed.

B The index data is deleted and recreated.

C The number of primary shards increases.

D The index becomes read-only.

Increasing number_of_replicas creates more copies of primary shards, which helps with fault tolerance and allows more nodes to serve search requests.

Which setting cannot be changed after index creation?

A refresh_interval

B number_of_shards

C number_of_replicas

D index.mapping.total_fields.limit

The number_of_shards is fixed when the index is created and cannot be changed later.

Why do we use shards in Elasticsearch?

A To encrypt data.

B To compress data.

C To backup data to external storage.

D To split data across nodes for scalability and performance.

Shards split the index data so Elasticsearch can store and search data efficiently across multiple nodes.

What is the default number_of_replicas in Elasticsearch?

A 1

B 2

C 0

D 3

By default, Elasticsearch creates 1 replica for each primary shard.

If an Elasticsearch node fails, what helps keep the data available?

A Index templates

B Primary shards only

C Replica shards

D Cluster settings

Replica shards provide copies of data so if a node with primary shards fails, replicas can serve the data.

Explain the roles of shards and replicas in Elasticsearch index settings.

Think about how data is stored and protected in Elasticsearch.

Describe how changing the number_of_replicas affects an Elasticsearch cluster.

Consider the benefits and costs of having more copies of data.