Async Hybrid Cache¶

Async Hybrid Cache helps async Python applications cache expensive work in local memory, optionally share values through Redis, and keep multiple application instances in sync with an invalidation bus.

flowchart LR
    App["Async app"] --> Cache["Async Hybrid Cache"]
    Cache --> L1["Local memory L1"]
    Cache -. optional .-> L2["Redis L2"]
    Cache -. optional .-> Bus["Invalidation bus"]
    Bus -. remove stale local keys .-> Peers["Other app instances"]

Quick Start¶

from async_hybrid_cache import CacheOptions, AsyncHybridCache

cache = AsyncHybridCache(
    options=CacheOptions(
        ttl_seconds=60,
        fail_safe_seconds=300,
        hard_timeout_seconds=5,
        jitter_seconds=5,
    ),
)

await cache.start()


@cache.cached(lambda user_id: f"user:{user_id}")
async def get_user(user_id: str) -> dict[str, str]:
    return {"id": user_id, "name": "Peter"}


user = await get_user("123")
await get_user.remove_cached("123")
await cache.stop()

Choose your path¶

Get a first cache working

Start with an in-memory cache, then add Redis when you need shared values.

Get started
Configure runtime behavior

Set TTLs, fail-safe stale reads, hard timeouts, and jitter for your app.

Configure cache policy
Run more than one app instance

Choose Redis Streams, RabbitMQ, Kafka, or PostgreSQL notifications for invalidation.

Choose an invalidation bus
Look up exact behavior

Check provider capabilities, decorator key behavior, options, and serializers.

Reference

Documentation map¶

Tutorials take you through a successful first result.
How-to guides solve a specific task in your application.
Reference lists exact options, defaults, and provider behavior.
Explanation describes the ideas behind the cache so you can make good decisions.