Why sharding is needed in MongoDB - Visual Breakdown

Concept Flow - Why sharding is needed

Start: Data grows large

↓

Single server struggles

↓

Performance drops

↓

Need to split data

↓

Implement sharding

↓

Data split across servers

↓

Improved performance & scalability

As data grows, one server can't handle it well, so we split data across servers using sharding to keep performance good.

Execution Sample

MongoDB

db.collection.insertMany([{_id:1,data:'A'}, {_id:2,data:'B'}, /* ... */])
// Data grows large
// Single server slows down
// Shard data across servers
// Queries go to correct shard

Shows data growing, single server slowing, then sharding data to improve query speed.

Execution Table

Step	Data Size	Server Load	Action	Result
1	Small	Low	Insert data	Fast insert and query
2	Medium	Moderate	Insert more data	Slightly slower queries
3	Large	High	Insert more data	Queries slow, server overloaded
4	Large	Overloaded	Decide to shard	Plan data split
5	Large	Distributed	Split data across shards	Load balanced, queries faster
6	Very Large	Distributed	Continue inserting	System scales well, performance stable

💡 Sharding distributes data to avoid overload and keep performance stable as data grows

Variable Tracker

Variable	Start	After Step 2	After Step 4	After Step 6
Data Size	Small	Medium	Large	Very Large
Server Load	Low	Moderate	Overloaded	Distributed
Query Speed	Fast	Slightly slower	Slow	Fast

Key Moments - 3 Insights

Why does the server load increase as data size grows?

Because more data means more work for the server to store and search, shown in execution_table steps 1 to 3.

Why can't we just keep adding data to one server?

One server has limits in memory, CPU, and disk speed, so performance drops as seen in step 3 where queries slow down.

How does sharding help improve performance?

Sharding splits data across servers, balancing load and making queries faster, as shown in steps 5 and 6.

Visual Quiz - 3 Questions

Look at the execution_table, what is the server load at step 3?

A High

B Moderate

C Low

D Distributed

💡 Hint

Check the 'Server Load' column in row for step 3 in execution_table

At which step does the system decide to shard the data?

A Step 2

B Step 4

C Step 3

D Step 6

💡 Hint

Look for the 'Action' column mentioning 'Decide to shard' in execution_table

If data size stayed small, what would happen to server load and query speed?

A Server load moderate, query speed slow

B Server load high, query speed slow

C Server load low, query speed fast

D Server load distributed, query speed fast

💡 Hint

Refer to step 1 in execution_table and variable_tracker for small data size

Concept Snapshot

Why Sharding is Needed:
- Data grows beyond single server capacity
- Server load increases, slowing queries
- Sharding splits data across servers
- Balances load and improves speed
- Enables system to scale with data size

Full Transcript

When data in a database grows large, a single server can struggle to handle all the data and queries. This causes the server load to increase and query speed to slow down. To fix this, we use sharding, which splits the data across multiple servers. This balances the load and keeps queries fast even as data grows. The execution table shows how data size and server load change step by step, and how sharding improves performance.