Functions

Browse Pinot function families and jump to the right execution-engine guidance.

Use this section as the entry point for Pinot SQL functions. The goal here is navigation: pick the function family, understand whether it is query-safe in the single-stage engine, and jump to the detailed page when you need full syntax or examples.

Most core function families work in both engines. Window functions require the multi-stage engine, and the engine column in the function index is the quickest way to confirm the supported execution model for a specific function.

Browse by family

Family	Typical use	Engine signal	Detailed docs
Aggregation	Metrics, distinct counts, sketches, percentiles	Both, with a few window-oriented helpers	Aggregation Functions
Array	Multi-value and array manipulation	Both	Array Functions
Binary	Hashing and encoding	Both	Binary Functions
DateTime	Bucketing, truncation, epoch conversion	Both	DateTime Functions
Funnel	Funnel analytics	Both	Funnel Functions
Geospatial	Spatial predicates and geometry conversion	Both	GeoSpatial Functions
Hash	Hash-based distribution and routing helpers	Both	Hash Functions
IP Address	Network and subnet checks	Both	IP Address Functions
JSON	Path extraction, formatting, and typed JSON access	Both, with query-safe typed variants	JSON Functions
Math	Arithmetic and numeric shaping	Both	Math Functions
Miscellaneous	Utility predicates and helpers	Both	Miscellaneous Functions
Null handling	Null-aware query behavior	Both, when null support is enabled	Null Handling Functions
Sketch	Approximate analytics helpers	Both	Sketch Functions
Statistical	Variance and covariance functions	Both	Statistical Functions
String	Case, slicing, tokenizing, regex, and formatting	Both	String Functions
Trigonometric	Sine, cosine, and related math	Both	Trigonometric Functions
Type conversion	Cast-like and conversion helpers	Both	Type Conversion Functions
UDFs	Custom extension points	Varies by implementation	User-Defined Functions
Vector / similarity	Embeddings and similarity search helpers	Both, where supported	Vector / Similarity Functions
Window	Ranking, framing, and rolling calculations	Multi-stage only	Window Functions

Ingestion-time transformations

If you are shaping records before they land in Pinot, use the curated transformation guide instead of starting from the raw function catalog.

Transformations

The transformation guide pulls together the functions most often used in schema updates, ingestion configs, and query projections, including typed JSON access, string shaping, date/time conversion, and multi-value helpers.

Engine choice

If you are deciding whether a query belongs on the single-stage engine or the multi-stage engine, use the decision guide first and then return here for the function family that matters.

SSE vs MSE

What this page covered

• The major Pinot function families and the engine signal for each family.

• Where to go for ingestion-time transformation guidance.

• Which engine decision guide to use before choosing a function-heavy query shape.

Next step

• Open the function family page for the syntax and examples you need, then confirm the engine constraints before writing the query.

Related pages

• SSE vs MSE

• Transformations

• Function Index