Database Partitioning

Database partitioning splits one large logical table into smaller physical pieces called partitions, typically by a date range, list of values, or hash of a key. The table still appears as a single object to queries.

Why it matters

Queries that filter on the partition key can skip irrelevant partitions entirely, which is called partition pruning. Partitioning also makes maintenance cheaper: dropping an old month of data is an instant operation instead of a huge delete. It differs from sharding, which spreads data across different servers, though the two ideas are often combined.

Example

-- PostgreSQL style
CREATE TABLE checks (
  ts timestamptz NOT NULL,
  site_id bigint,
  response_ms int
) PARTITION BY RANGE (ts);

CREATE TABLE checks_2026_10 PARTITION OF checks
  FOR VALUES FROM ('2026-10-01') TO ('2026-11-01');

Common mistakes

• Partitioning a small table, which adds overhead without benefit.
• Writing queries that do not filter on the partition key, so every partition is scanned.
• Mixing up partitioning (one server) with sharding (many servers).

Related

• Database sharding
• Time-series database
• Database index

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.