Shard sizing strategy in Elasticsearch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the recommended shard size for optimal performance?

Given the following Elasticsearch shard sizing guidelines, what is the recommended shard size range for optimal performance?

Options are in gigabytes (GB).

A 1-5 GB

B 10-50 GB

C 100-200 GB

D 500-1000 GB

💡 Hint

Think about balancing shard size to avoid overhead and ensure fast recovery.

Elasticsearch recommends shard sizes between 10 and 50 GB for optimal performance. Smaller shards increase overhead, while larger shards slow down recovery and search.

🧠 Conceptual

intermediate

Why avoid too many small shards in Elasticsearch?

What is the main reason to avoid having too many small shards in an Elasticsearch cluster?

A Small shards cause excessive overhead and resource consumption

B Small shards improve search speed significantly

C Small shards increase disk space usage

D Small shards reduce cluster stability

💡 Hint

Think about how Elasticsearch manages shards internally.

Too many small shards cause overhead in managing them, increasing CPU and memory usage, which reduces cluster efficiency.

❓ Predict Output

advanced

What happens if shard size exceeds recommended limits?

Consider an Elasticsearch index with shards sized around 200 GB each. What is the most likely impact on cluster behavior?

A No impact on cluster performance

B Faster indexing and search performance

C Reduced disk usage and faster backups

D Slower shard recovery and increased search latency

💡 Hint

Think about how large shards affect recovery and search operations.

Large shards slow down recovery times and increase search latency because more data must be processed per shard.

🔧 Debug

advanced

Identify the shard sizing mistake in this scenario

An Elasticsearch cluster has 100 indices, each with 20 shards sized about 1 GB. The cluster is experiencing high CPU usage and slow queries. What is the likely shard sizing mistake?

A Too many small shards causing overhead

B Shards are too large causing slow recovery

C Insufficient number of shards per index

D Shard size is optimal; issue is unrelated

💡 Hint

Consider how shard count and size affect resource usage.

Having many small shards increases overhead, leading to high CPU usage and slower queries.

🚀 Application

expert

Calculate total shard size and count for a 5 TB index

You have a 5 TB Elasticsearch index. You want to keep shard sizes between 20 GB and 40 GB for optimal performance. How many shards should you create?

Choose the correct shard count range.

A 500 to 1000 shards

B 10 to 25 shards

C 125 to 250 shards

D 50 to 100 shards

💡 Hint

Divide total index size by shard size range to find shard count.

5 TB = 5000 GB. Dividing by 40 GB gives 125 shards minimum, dividing by 20 GB gives 250 shards maximum.