Deployment

Choose the right Redis topology for your availability and scalability needs.

Topology Comparison

Topology	HA	Scalability	Complexity	Use Case
Single	No	Vertical	Low	Dev, small apps
Sentinel	Yes	Vertical	Medium	Production HA
Cluster	Yes	Horizontal	High	Large scale

Single Node

Best for: Development, small applications, non-critical workloads

RedisModule.forRoot({
  clients: {
    host: 'redis',
    port: 6379,
  },
})

Limitations:

• No automatic failover
• Memory limited to single machine
• Single point of failure

Sentinel (Recommended for HA)

Best for: Production with high availability needs

RedisModule.forRoot({
  clients: {
    sentinels: [
      { host: 'sentinel-1', port: 26379 },
      { host: 'sentinel-2', port: 26379 },
      { host: 'sentinel-3', port: 26379 },
    ],
    name: 'mymaster',
    password: process.env.REDIS_PASSWORD,
  },
})

Failover behavior:

1. Sentinel detects master failure
2. Sentinels elect new master
3. Client redirected to new master
4. ~30 seconds failover time

Cluster

Best for: Large-scale applications needing horizontal scaling

// ioredis cluster mode
import { Cluster } from 'ioredis';

const cluster = new Cluster([
  { host: 'node-1', port: 6379 },
  { host: 'node-2', port: 6379 },
  { host: 'node-3', port: 6379 },
], {
  redisOptions: {
    password: process.env.REDIS_PASSWORD,
  },
  scaleReads: 'slave',
});

Key hashing:

• Keys distributed across 16384 slots
• Use {hash_tag} to co-locate related keys
• Some commands limited to single slot

// These keys will be on same shard
'user:{123}:profile'
'user:{123}:settings'
'user:{123}:orders'

Decision Matrix

Factor	Single	Sentinel	Cluster
Data size	<10GB	<100GB	>100GB
Availability needs	Low	High	High
Operations team	Small	Medium	Large
Budget	Low	Medium	Higher
Failover time	Manual	~30s	~15s

Cloud Managed Options

Provider	Service	Topology Support
AWS	ElastiCache	Single, Cluster
GCP	Memorystore	Single, Cluster
Azure	Azure Cache	Single, Cluster
Redis Labs	Redis Cloud	All

Benefits of managed:

• Automatic patching
• Backup/restore
• Monitoring included
• Simplified operations

High Availability Architecture

Persistence Configuration

Mode	Durability	Performance
None	Lowest	Highest
RDB	Medium	High
AOF	High	Medium
RDB + AOF	Highest	Lower

For caching: RDB or none For queues/locks: AOF recommended

# redis.conf
appendonly yes
appendfsync everysec

Next Steps

• Tuning — Performance optimization
• Connection Management — Connection config