redis-best-practices

Redis Best Practices

Core Principles

• Use Redis for caching, session storage, real-time analytics, and message queuing
• Choose appropriate data structures for your use case
• Implement proper key naming conventions and expiration policies
• Design for high availability and persistence requirements
• Monitor memory usage and optimize for performance

Key Naming Conventions

• Use colons as namespace separators
• Include object type and identifier in key names
• Keep keys short but descriptive
• Use consistent naming patterns across your application

# Good key naming examplesuser:1234:profileuser:1234:sessionsorder:5678:itemscache:api:products:listqueue:email:pendingsession:abc123def456rate_limit:api:user:1234

Data Structures

Strings

• Use for simple key-value storage, counters, and caching
• Consider using MGET/MSET for batch operations

# Simple cachingSET cache:user:1234 '{"name":"John","email":"[email protected]"}' EX 3600# CountersINCR stats:pageviews:homepageINCRBY stats:downloads:file123 5# Atomic operationsSETNX lock:resource:456 "owner:abc" EX 30

Hashes

• Use for objects with multiple fields
• More memory-efficient than multiple string keys
• Supports partial updates

# Store user profileHSET user:1234 name "John Doe" email "[email protected]" created_at "2024-01-15"# Get specific fieldsHGET user:1234 emailHMGET user:1234 name email# Increment numeric fieldsHINCRBY user:1234 login_count 1# Get all fieldsHGETALL user:1234

Lists

• Use for queues, recent items, and activity feeds
• Consider blocking operations for queue consumers

# Message queueLPUSH queue:emails '{"to":"[email protected]","subject":"Welcome"}'RPOP queue:emails# Blocking pop for workersBRPOP queue:emails 30# Recent activity (keep last 100)LPUSH user:1234:activity "viewed product 567"LTRIM user:1234:activity 0 99# Get recent itemsLRANGE user:1234:activity 0 9

Sets

• Use for unique collections, tags, and relationships
• Supports set operations (union, intersection, difference)

# User tags/interestsSADD user:1234:interests "technology" "music" "travel"# Check membershipSISMEMBER user:1234:interests "music"# Find common interestsSINTER user:1234:interests user:5678:interests# Online users trackingSADD online:users "user:1234"SREM online:users "user:1234"SMEMBERS online:users

Sorted Sets

• Use for leaderboards, priority queues, and time-series data
• Elements sorted by score

# LeaderboardZADD leaderboard:game1 1500 "player:123" 2000 "player:456" 1800 "player:789"# Get top 10ZREVRANGE leaderboard:game1 0 9 WITHSCORES# Get player rankZREVRANK leaderboard:game1 "player:123"# Time-based data (score = timestamp)ZADD events:user:1234 1705329600 "login" 1705330000 "purchase"# Get events in time rangeZRANGEBYSCORE events:user:1234 1705329600 1705333200

Streams

• Use for event streaming and log data
• Supports consumer groups for distributed processing

# Add events to streamXADD events:orders * customer_id 1234 product_id 567 amount 99.99# Read from streamXREAD COUNT 10 STREAMS events:orders 0# Consumer groupsXGROUP CREATE events:orders order-processors $ MKSTREAMXREADGROUP GROUP order-processors worker1 COUNT 10 STREAMS events:orders ># Acknowledge processed messagesXACK events:orders order-processors 1234567890-0

Caching Patterns

Cache-Aside Pattern

# Pseudo-code for cache-asidedef get_user(user_id):    # Try cache first    cached = redis.get(f"cache:user:{user_id}")    if cached:        return json.loads(cached)    # Cache miss - fetch from database    user = database.get_user(user_id)    # Store in cache with expiration    redis.setex(f"cache:user:{user_id}", 3600, json.dumps(user))    return user

Write-Through Pattern

def update_user(user_id, data):    # Update database    database.update_user(user_id, data)    # Update cache    redis.setex(f"cache:user:{user_id}", 3600, json.dumps(data))

Cache Invalidation

# Delete specific cacheDEL cache:user:1234# Delete by pattern (use with caution in production)# Use SCAN instead of KEYS for large datasetsSCAN 0 MATCH cache:user:* COUNT 100# Tag-based invalidation using setsSADD cache:tags:user:1234 "cache:user:1234:profile" "cache:user:1234:orders"# Invalidate all related cachesSMEMBERS cache:tags:user:1234# Then delete each key

Expiration and Memory Management

TTL Best Practices

• Always set TTL on cache keys
• Use jitter to prevent thundering herd
• Consider sliding expiration for session data

# Set with expirationSET cache:data:123 "value" EX 3600# Set expiration on existing keyEXPIRE cache:data:123 3600# Check TTLTTL cache:data:123# Persist key (remove expiration)PERSIST cache:data:123

Memory Management

# Check memory usageINFO memory# Get key memory usageMEMORY USAGE cache:large:object# Configure max memory policyCONFIG SET maxmemory 2gbCONFIG SET maxmemory-policy allkeys-lru

Transactions and Atomicity

MULTI/EXEC Transactions

# Transaction blockMULTIINCR stats:viewsLPUSH recent:views "page:123"EXEC# Watch for optimistic lockingWATCH user:1234:balancebalance = GET user:1234:balanceMULTISET user:1234:balance (balance - 100)EXEC

Lua Scripts

• Use for complex atomic operations
• Scripts execute atomically

-- Rate limiting scriptlocal key = KEYS[1]local limit = tonumber(ARGV[1])local window = tonumber(ARGV[2])local current = tonumber(redis.call('GET', key) or '0')if current >= limit then    return 0endredis.call('INCR', key)if current == 0 then    redis.call('EXPIRE', key, window)endreturn 1

# Execute Lua scriptEVAL "return redis.call('GET', KEYS[1])" 1 mykey

Pub/Sub and Messaging

# PublisherPUBLISH channel:notifications '{"type":"alert","message":"New order"}'# SubscriberSUBSCRIBE channel:notifications# Pattern subscriptionPSUBSCRIBE channel:*

High Availability

Replication

• Use replicas for read scaling
• Configure proper persistence on master

# On replicaREPLICAOF master_host 6379# Check replication statusINFO replication

Redis Sentinel

• Use for automatic failover
• Deploy at least 3 Sentinel instances

Redis Cluster

• Use for horizontal scaling
• Data automatically sharded across nodes
• Use hash tags for related keys

# Hash tags ensure keys go to same slotSET {user:1234}:profile "data"SET {user:1234}:settings "data"

Persistence

RDB Snapshots

# Manual snapshotBGSAVE# Configure automatic snapshotsCONFIG SET save "900 1 300 10 60 10000"

AOF (Append-Only File)

# Enable AOFCONFIG SET appendonly yesCONFIG SET appendfsync everysec# Rewrite AOFBGREWRITEAOF

Security

• Require authentication
• Use TLS for connections
• Bind to specific interfaces
• Disable dangerous commands

# Set passwordCONFIG SET requirepass "your_strong_password"# AuthenticateAUTH your_strong_password# Rename dangerous commands (in redis.conf)rename-command FLUSHALL ""rename-command FLUSHDB ""rename-command KEYS ""

Monitoring

# Server infoINFO# Memory statsINFO memory# Client connectionsCLIENT LIST# Slow logSLOWLOG GET 10# Monitor commands (debug only)MONITOR# Key count per databaseINFO keyspace

Connection Management

• Use connection pooling
• Set appropriate timeouts
• Handle reconnection gracefully

# Python example with connection poolimport redispool = redis.ConnectionPool(    host='localhost',    port=6379,    max_connections=50,    socket_timeout=5,    socket_connect_timeout=5)redis_client = redis.Redis(connection_pool=pool)

Performance Tips

• Use pipelining for batch operations
• Avoid large keys (>100KB values)
• Use SCAN instead of KEYS in production
• Monitor and optimize memory usage
• Consider using RedisJSON for complex JSON operations

# Pipeline example (pseudo-code)pipe = redis.pipeline()pipe.get("key1")pipe.get("key2")pipe.set("key3", "value")results = pipe.execute()