Hashed Sharding in MongoDB

Hashed sharding in MongoDB distributes data using the hashed value of a shard key to balance load and improve scalability.

• Evenly distributes data and queries across shards, preventing hotspots from skewed or sequential values.
• Improves read and write performance at scale through balanced load.
• Stores data as chunks distributed across multiple shard servers.

Hashed Shard Key

Hashed sharding distributes data across a distributed database using the hashed value of a shard key to ensure balanced load and avoid hotspots.

• Uses a hash function on the shard key to evenly distribute data.
• Prevents hotspots by spreading reads and writes across shards.
• Shard key choice is critical for balanced distribution and performance.
• Requires analyzing access patterns, queries, and data characteristics.

1. Sharding on a Single Field Hashed Index

Sharding on a single-field hashed index distributes documents across shards using the hashed value of one field, helping balance load and improve scalability, especially for write-heavy workloads.

• Uses a hashed value of a single field as the shard key.
• Evenly distributes data to prevent hotspots.
• Improves write scalability and load balancing.
• Simple shard key design and setup.
• Not suitable for range queries or ordered queries on the shard key.

2. Sharding on a Compound Hashed Index

Sharding on a compound hashed index distributes data using the combined hashed value of multiple fields, offering more flexible query support with balanced data distribution.

• Uses a compound shard key with one hashed field and additional non-hashed fields.
• Supports complex, multi-criteria query patterns.
• Evenly distributes related data across shards for balanced load.
• Requires careful shard key design to avoid uneven distribution.

Implementing Hashed Sharding

An example of implementing hashed sharding in MongoDB.

Step 1: Enable Sharding

Before enabling sharding on a collection, ensure that the MongoDB deployment is configured for sharding.

# Enable sharding on the database
sh.enableSharding("mydatabase")
# Enable sharding on the collection with a specified shard key
sh.shardCollection("mydatabase.mycollection", { "myShardKeyField": "hashed" })

Step 2: Insert Data

Insert data into the sharded collection. MongoDB will automatically distribute documents across shards based on the hashed shard key.

db.mycollection.insertOne({
  "name": "John Doe",
  "age": 30,
  "myShardKeyField": "someValue"
})

Step 3: Query Sharded Data

Query data from the sharded collection. MongoDB will route queries to the appropriate shards based on the hashed shard key.

db.mycollection.find({ "myShardKeyField": "someValue" })

Example: We have a sharded collection named "mycollection" with hashed sharding on the "myShardKeyField" field, querying the data will produce output similar to the following:

{
  "_id": ObjectId("60f9d7ac345b7c9df348a86e"),
  "name": "John Doe",
  "age": 30,
  "myShardKeyField": "someValue"
}

Advantages of Hashed Sharding

Hashed sharding offers several benefits:

• Even Data Distribution: Hash values spread data evenly across shards, preventing hotspots and skewed load.
• Predictable Shard Placement: Hashing provides consistent shard assignment for documents.
• Write Scalability: Insert operations are distributed across shards, improving write throughput.
• Balanced Read/Write Load: Workloads are evenly shared across the cluster.
• Reduced Hotspot Risk: Sequential or skewed keys don’t overload a single shard.

Hashed Vs Ranged Sharding