Skip to main content

Geospatial Indexes

Redis provides built-in geospatial capabilities that allow you to:

  • Store geographic locations (longitude, latitude)
  • Query nearby points within a radius
  • Calculate distances between two locations
  • Sort by distance
  • Use bounding boxes and circular radius queries

Redis geospatial features are built on top of the sorted set (ZSET) data structure using a technique called Geohash encoding.

How Redis Geospatial Works Internally​

  1. Every geospatial point is stored as a Geohash, which is a 52-bit integer.
  2. Redis stores that integer inside a sorted set.
  3. Geospatial commands (GEOADD, GEORADIUS, etc.) handle encoding and querying.

Redis does not store altitude or 3D coordinates—just latitude & longitude.

Redis Geospatial Commands​

CommandDescription
GEOADDAdd a location with longitude, latitude, and name
GEOPOSGet coordinates of a member
GEODISTGet distance between two points
GEORADIUSFind members within a radius (circle)
GEORADIUSBYMEMBERRadius query based on another member
GEORADIUS_RO, GEORADIUSBYMEMBER_RORead-only versions
GEOHASHGet geohash strings

  1. Adding Geospatial Data

    GEOADD cities -74.00597 40.71427 "NewYork"
    • cities = key name holding geospatial index
    • Each entry requires: longitude, latitude, member name
    • Redis converts data → geohash → stores in sorted set

  2. Retrieve Coordinates of a City

    GEOPOS cities "NewYork"

    Output:

    1) 1) "-74.00596940505599976"
       2) "40.71426910448555265"

    Redis returns the actual stored coordinates (with high precision).

  3. Get Distance Between Cities

    GEODIST cities "NewYork" "Chicago" km

    Output

    1146.3023

    Redis uses spherical Earth approximation (Haversine formula).

  4. Find Nearby Locations (Radius Query)

    Example: Find cities within 1500 km of New York

    GEORADIUS cities -74.00597 40.71427 1500 km

    Output:

    1) "Chicago"
    1) "NewYork"
    • Center → given manually (longitude/latitude)
    • Radius → 1500 km
    • Returns matching cities sorted by distance from center (closest first)

  5. Radius Query Based on Another

    Example: Cities within 1800 km of Chicago

    GEORADIUSBYMEMBER cities "Chicago" 1800 km

    Output:

    1) "NewYork"
    1) "Chicago"
    • Uses Chicago’s coordinates as the center automatically
    • Very useful for “find nearby users”, “find nearby restaurants”, etc.

  6. Radius Query with Extra Details

    Example: Include coordinates + distance + geohash

    GEORADIUS cities -74 40 1500 km WITHCOORD WITHDIST WITHHASH

    Output:

    1) 1) "NewYork"
       2) "14.1"                 ← distance
       3) "3478007285230310"     ← geohash
       4) 1) "-74.0059"          ← longitude
          2) "40.7142"           ← latitude

    Additional flags:

    • WITHDIST → distance from center
    • WITHCOORD → return lat/lon
    • WITHHASH → internal hash used in Redis

  7. Get Geohash Values

    GEOHASH cities "NewYork"

    Output:

    1) "dr5regw3pg0"

    Geohashes provide:

    • Location encoding
    • Prefix similarity → spatial proximity (Locations close on map share geohash prefixes)