Vertical vs Horizontal Database Sharding: Benefits, Drawbacks, and When to Use Each

Vertical Sharding

Vertical sharding imagines a database as many large "data blocks" (tables) that are cut vertically and dispersed across multiple database hosts.

Advantages

Database splitting is straightforward and rules are clear.

Application modules are well defined, making integration easy.

Data maintenance is convenient and locating data is simple.

Disadvantages

Some table relationships cannot be handled at the database level and must be resolved in application code.

For extremely frequent access and massive tables, performance bottlenecks may remain.

Transaction processing becomes complex.

Scalability is limited after a certain degree of splitting.

Over‑splitting can make the system overly complex and hard to maintain.

Horizontal Sharding

Horizontal sharding distributes a highly accessed table across multiple tables according to a rule based on a specific field, with each table containing a subset of the data.

Advantages

Table relationships can largely be resolved at the database side.

Performance bottlenecks for ultra‑large or high‑load tables are avoided.

Overall application architecture changes are relatively minor.

Transaction handling is comparatively simple.

As long as the partition rule is well defined, scalability limits are rare.

Disadvantages

Partition rules are relatively complex and hard to abstract for the entire database.

Later data maintenance becomes more difficult, with manual data location harder.

Higher coupling between application modules may hinder future data migration or further splitting.