cloudflare-kv

Cloudflare Workers KV enables edge storage by binding a KV namespace to your Worker and using get/put/delete operations. Data stored in KV is replicated globally across Cloudflare's network, delivering low-latency reads and writes from edge locations closest to your users.

Yes, KV supports per-key TTL (time-to-live) settings. When you put a key, specify the ttl parameter in seconds to automatically expire that key. You can also set cacheTtl to control how long responses are cached at the edge.

KV's list operation supports cursor-based pagination and prefix filtering. Use the cursor returned from a list call to fetch the next batch of keys, and filter results by prefix to narrow your dataset before paginating.

KV allows you to store custom metadata alongside each key-value pair. Store metadata as JSON or structured data in the value itself, or use metadata parameters to track creation time, expiration, or application-specific attributes for retrieval.

KV supports bulk operations for reading and writing multiple keys in a single request. Use batch get/put methods to reduce round trips and improve performance when working with configuration data, feature flags, or state across your Workers.

KV is eventually consistent across regions with eventual consistency guarantees. Be aware of rate limits, namespace quotas, and the distinction between development and production bindings when architecting cache invalidation and error handling for high-traffic scenarios.