Cache Connectors

The Cache subsystem provides high-performance temporary data storage for frequently accessed information. It supports TTL (Time-To-Live) expiration, automatic cleanup, and various storage backends for different performance and scalability requirements.

Available Connectors

RAM

Role: In-memory cache connector
Summary: Provides ultra-fast caching using process memory. Ideal for single-node applications requiring maximum performance with automatic memory management.

Setting	Type	Required	Default	Description
No specific settings	any	No	-	RAMCache accepts any settings object

Example Configuration:

import { SRE } from '@smythos/sre';

SRE.init({
    Cache: {
        Connector: 'RAM',
        Settings: {},
    },
});
Copy

Use Cases:

• Single-node applications
• Development and testing
• Ultra-low latency requirements
• Session storage
• Temporary computation results

---

Redis

Role: Redis-based distributed cache connector
Summary: Provides scalable, distributed caching using Redis server. Supports clustering, persistence, and advanced data structures for high-performance applications.

Setting	Type	Required	Default	Description
name	string	Yes	-	Redis master name (for Sentinel mode)
password	string	Yes	-	Redis authentication password
hosts	string | string[] | any[]	Yes	-	Redis host(s) - single host string or array for Sentinel
prefix	string	No	-	Key prefix for namespace isolation

Example Configuration:

import { SRE } from '@smythos/sre';

SRE.init({
    Cache: {
        Connector: 'Redis',
        Settings: {
            name: 'mymaster',
            password: process.env.REDIS_PASSWORD,
            hosts: 'localhost:6379',
            prefix: 'myapp:cache:',
        },
    },
});
Copy

Sentinel Configuration:

import { SRE } from '@smythos/sre';

SRE.init({
    Cache: {
        Connector: 'Redis',
        Settings: {
            name: 'mymaster',
            password: process.env.REDIS_PASSWORD,
            hosts: ['sentinel1:26379', 'sentinel2:26379', 'sentinel3:26379'],
        },
    },
});
Copy

Use Cases:

• Multi-node distributed applications
• Production environments requiring persistence
• Session sharing across instances
• High-availability caching
• Advanced data structure operations

---

LocalStorage

Role: File-based cache connector
Summary: Provides persistent caching using local filesystem storage. Suitable for applications requiring cache persistence across restarts without external dependencies.

Setting	Type	Required	Default	Description
folder	string	No	~/.smyth/storage	Directory path for cache files

Example Configuration:

import { SRE } from '@smythos/sre';

SRE.init({
    Cache: {
        Connector: 'LocalStorage',
        Settings: {
            folder: './data/cache',
        },
    },
});
Copy

Use Cases:

• Single-node applications requiring persistence
• Development environments
• Applications with intermittent connectivity
• Cost-sensitive deployments
• Simple deployment scenarios

---

S3

Role: Amazon S3-based cache connector
Summary: Provides cloud-based caching using Amazon S3 storage. Offers unlimited scalability with built-in durability and global accessibility.

Setting	Type	Required	Default	Description
bucketName	string	Yes	-	S3 bucket name for cache storage
region	string	Yes	-	AWS region of the bucket
accessKeyId	string	Yes	-	AWS access key ID
secretAccessKey	string	Yes	-	AWS secret access key

Example Configuration:

import { SRE } from '@smythos/sre';

SRE.init({
    Cache: {
        Connector: 'S3',
        Settings: {
            bucketName: 'my-app-cache',
            region: 'us-east-1',
            accessKeyId: process.env.AWS_ACCESS_KEY_ID,
            secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY,
        },
    },
});
Copy

Use Cases:

• Multi-region distributed applications
• Serverless architectures
• Applications requiring unlimited cache storage
• Long-term cache persistence
• Integration with AWS ecosystem

Performance Notes:

• S3 cache has higher latency than memory/Redis-based solutions
• Best suited for larger objects or infrequently accessed data
• Consider using CloudFront for global distribution
• Monitor S3 costs for high-frequency cache operations

Configuration Notes

Redis Configuration

• When hosts is a single string, it's treated as a direct Redis connection
• When hosts is an array, it's treated as Redis Sentinel configuration
• The name field is required for Sentinel mode to identify the master
• Password can also be provided via REDIS_PASSWORD environment variable
• Hosts can also be provided via REDIS_HOSTS environment variable

Environment Variables

The Redis connector supports these environment variables as fallbacks:

• REDIS_HOSTS - Redis host(s) configuration
• REDIS_PASSWORD - Redis authentication password
• REDIS_MASTER_NAME - Redis master name for Sentinel mode