Skip to main content
Redis geospatial indexes allow you to store longitude/latitude coordinates and perform radius queries, distance calculations, and proximity searches. Built on sorted sets using geohash encoding.

​
Use Cases

  • Location-based services: Find nearby restaurants, stores, or users
  • Delivery tracking: Find drivers near a delivery address
  • Ride-sharing: Match riders with nearby drivers
  • Real estate: Search properties within an area
  • Gaming: Find players in proximity
  • IoT devices: Locate sensors or assets in a region

​
Key Commands

​
Adding Locations

# Add single location (longitude, latitude, member)
redis> GEOADD locations -122.4194 37.7749 "San Francisco"
(integer) 1

# Add multiple locations
redis> GEOADD locations 
  -118.2437 34.0522 "Los Angeles" 
  -73.9352 40.7306 "Brooklyn" 
  -73.9949 40.7282 "Manhattan"
(integer) 3

# Update existing location (XX option)
redis> GEOADD locations XX -122.4200 37.7750 "San Francisco"
(integer) 0

# Add only if doesn't exist (NX option)
redis> GEOADD locations NX -122.3321 47.6062 "Seattle"
(integer) 1

​
Radius Queries

# Find locations within radius
redis> GEORADIUS locations -122.4194 37.7749 100 km
1) "San Francisco"

# Find with distances
redis> GEORADIUS locations -122.4194 37.7749 500 km WITHDIST
1) 1) "San Francisco"
   2) "0.0002"
2) 1) "Los Angeles"
   2) "559.2362"

# Find with coordinates
redis> GEORADIUS locations -122.4194 37.7749 500 km WITHCOORD
1) 1) "San Francisco"
   2) 1) "-122.41939842700958252"
      2) "37.77490009169188668"

# Radius from existing member
redis> GEORADIUSBYMEMBER locations "San Francisco" 500 km
1) "San Francisco"
2) "Los Angeles"

# Limit results and sort
redis> GEORADIUS locations -122.4194 37.7749 1000 km COUNT 2 ASC
1) "San Francisco"
2) "Los Angeles"

​
Modern GEOSEARCH (Redis 6.2+)

# Search by radius from coordinates
redis> GEOSEARCH locations FROMLONLAT -122.4194 37.7749 BYRADIUS 500 km
1) "San Francisco"
2) "Los Angeles"

# Search by radius from member
redis> GEOSEARCH locations FROMMEMBER "San Francisco" BYRADIUS 500 km
1) "San Francisco"
2) "Los Angeles"

# Search in bounding box (width x height)
redis> GEOSEARCH locations FROMLONLAT -122.4194 37.7749 BYBOX 600 600 km
1) "San Francisco"
2) "Los Angeles"

# With distances and coordinates
redis> GEOSEARCH locations FROMMEMBER "San Francisco" BYRADIUS 500 km 
  WITHDIST WITHCOORD COUNT 5 ASC

​
Distance and Position

# Get distance between locations
redis> GEODIST locations "San Francisco" "Los Angeles" km
"559.2362"

# Get in different units
redis> GEODIST locations "San Francisco" "Los Angeles" mi
"347.4210"

# Get coordinates
redis> GEOPOS locations "San Francisco" "Brooklyn"
1) 1) "-122.41939842700958252"
   2) "37.77490009169188668"
2) 1) "-73.93519818782806396"
   2) "40.73059881448093077"

# Get geohash
redis> GEOHASH locations "San Francisco"
1) "9q8yyk8yuv0"

​
Time Complexity

CommandTime ComplexityDescription
GEOADDO(log N)Add location
GEORADIUSO(N+log M)N=radius items, M=total
GEORADIUSBYMEMBERO(N+log M)N=radius items, M=total
GEOSEARCHO(N+log M)N=box items, M=total
GEODISTO(log N)Distance between two
GEOPOSO(N)N=members requested
GEOHASHO(N)N=members requested

​
Patterns and Examples

​
Find Nearby Restaurants

# Add restaurants with coordinates
redis> GEOADD restaurants 
  -122.4183 37.7799 "Tartine Bakery" 
  -122.4161 37.7833 "Zuni Cafe" 
  -122.4089 37.7835 "State Bird Provisions"
(integer) 3

# Find restaurants within 2km
redis> GEOSEARCH restaurants 
  FROMLONLAT -122.4194 37.7749 
  BYRADIUS 2 km 
  WITHDIST 
  ASC
1) 1) "Tartine Bakery"
   2) "0.5634"
2) 1) "Zuni Cafe"
   2) "0.9876"
3) 1) "State Bird Provisions"
   2) "1.2345"

# Find 5 closest restaurants
redis> GEOSEARCH restaurants 
  FROMMEMBER "Tartine Bakery" 
  BYRADIUS 5 km 
  COUNT 5 
  ASC

​
Ride-Sharing Driver Matching

# Add available drivers
redis> GEOADD drivers:available 
  -122.4183 37.7799 "driver:123" 
  -122.4089 37.7835 "driver:456" 
  -122.4250 37.7694 "driver:789"
(integer) 3

# Find nearest driver for pickup
redis> GEOSEARCH drivers:available 
  FROMLONLAT -122.4194 37.7749 
  BYRADIUS 3 km 
  COUNT 1 
  ASC 
  WITHDIST
1) 1) "driver:789"
   2) "0.6523"

# Remove driver when assigned
redis> ZREM drivers:available "driver:789"
(integer) 1

# Calculate distance to destination
redis> GEODIST drivers:available "driver:123" "dropoff_location" km
"4.5678"

​
Store Locator

# Add store locations
redis> GEOADD stores 
  -122.4194 37.7749 "store:sf:union" 
  -122.3321 47.6062 "store:seattle:downtown" 
  -118.2437 34.0522 "store:la:santa-monica"
(integer) 3

# Find stores in customer's area
redis> GEOSEARCH stores 
  FROMLONLAT -122.4000 37.7800 
  BYRADIUS 5 km 
  WITHCOORD 
  WITHDIST
1) 1) "store:sf:union"
   2) "2.3456"
   3) 1) "-122.41939842700958252"
      2) "37.77490009169188668"

# Check if store is within delivery zone (10km)
redis> GEODIST stores "store:sf:union" "customer_location" km
"3.4567"  # Yes, within 10km

# Add properties
redis> GEOADD properties 
  -122.4194 37.7749 "property:1001" 
  -122.4183 37.7799 "property:1002" 
  -122.4089 37.7835 "property:1003"
(integer) 3

# Search in rectangular area (bounding box)
redis> GEOSEARCH properties 
  FROMLONLAT -122.4194 37.7749 
  BYBOX 2 2 km 
  WITHCOORD
1) 1) "property:1001"
   2) 1) "-122.41939842700958252"
      2) "37.77490009169188668"
2) 1) "property:1002"
   2) 1) "-122.41829872131347656"
      2) "37.77989959716796875"

​
Gaming - Find Nearby Players

# Update player positions
redis> GEOADD players 
  -122.4183 37.7799 "player:alice" 
  -122.4089 37.7835 "player:bob"
(integer) 2

# Find players within 1km
redis> GEOSEARCH players 
  FROMMEMBER "player:alice" 
  BYRADIUS 1 km 
  WITHDIST
1) 1) "player:alice"
   2) "0.0000"
2) 1) "player:bob"
   2) "0.8765"

# Update position when player moves
redis> GEOADD players -122.4150 37.7820 "player:alice"
(integer) 0

​
Geohash Encoding

Redis uses geohash to encode coordinates: 
# Get geohash
redis> GEOHASH locations "San Francisco"
1) "9q8yyk8yuv0"

# Nearby locations have similar prefixes
redis> GEOHASH locations "San Francisco" "Los Angeles"
1) "9q8yyk8yuv0"  # San Francisco
2) "9q5ctr4xg80"  # Los Angeles
# Common prefix: "9q" (both in California)
Geohashes with longer common prefixes are geographically closer. Redis stores these as sorted set scores, enabling efficient range queries.

​
Units

Supported distance units:
  • m: meters
  • km: kilometers
  • mi: miles
  • ft: feet
redis> GEODIST locations "San Francisco" "Los Angeles" km
"559.2362"
redis> GEODIST locations "San Francisco" "Los Angeles" mi
"347.4210"

​
Internal Structure

Geospatial indexes are stored as sorted sets: 
# Geospatial index is a sorted set
redis> TYPE locations
zset

# Can use sorted set commands
redis> ZRANGE locations 0 -1
1) "Los Angeles"
2) "San Francisco"
3) "Brooklyn"
4) "Manhattan"

# Remove location
redis> ZREM locations "Manhattan"
(integer) 1

# Get all locations with scores (geohashes)
redis> ZRANGE locations 0 -1 WITHSCORES

​
Best Practices

  1. Use GEOSEARCH instead of deprecated GEORADIUS/GEORADIUSBYMEMBER
  2. Set expiration on temporary location data (active drivers, players)
  3. Use COUNT to limit results for performance
  4. Index by category (restaurants, hotels, stores) for faster queries
  5. Update locations when entities move
GEORADIUS and GEORADIUSBYMEMBER are deprecated since Redis 6.2. Use GEOSEARCH instead for better performance and more features.
For applications with many categories (restaurants, hotels, stores), create separate geospatial indexes per category rather than one large index. This significantly improves query performance.

​
Limitations

  • Earth assumed as sphere: Slight inaccuracy at poles
  • Coordinate precision: ±0.5 meters
  • No altitude: Only 2D coordinates (longitude, latitude)
  • Periodic updates: Moving entities require position updates

​
Next Steps

Sorted Sets

Underlying data structure

Sorted Set Commands

For advanced operations

Build docs developers (and LLMs) love

Get started for freeTalk to us