Managed Valkey on Zenifra
Managed Valkey provides a Redis-compatible instance for three different needs: key-value data, caching, and work queues. The card selected in the Console defines the profile; there is no later profile selector.
Choose the right product
| Product | Use it for | Storage | Commercial plans |
|---|---|---|---|
| Key‑Value | sessions, shared state, counters, and locks | Included and durable | db-free, db-starter, db-basic, db-premium, db-enterprise |
| Cache | computed responses, queries, and disposable data | Non-durable | cache-free, cache-starter, cache-basic, cache-premium, cache-enterprise |
| Queues | asynchronous tasks, workers, and durable events | Included and durable | queue-free, queue-starter, queue-basic, queue-premium, queue-enterprise |
Prices, benefits, capacity, and billing mode should be read from the catalog or the Console cards. They may change without requiring application code changes.
Security and connection
The endpoint is publicly reachable over IPv4, but TLS and authentication are required. The initial connection uses the valkeys:// scheme and the default username; the full password is revealed only during project creation or credential rotation.
export VALKEY_URL='valkeys://default:[email protected]:30000/0'Store VALKEY_URL in a secret manager or protected variable. Never put it in source code, logs, tickets, or images. Some clients do not recognize valkeys://; in that case, read the host, port, username, and password and configure TLS explicitly, or replace only the scheme with rediss://. Never disable certificate or hostname validation.
See Connect to Valkey for examples in eight languages and Rotate credentials for the API operation.
Choosing a profile
- Choose Key‑Value when data is part of application state and must survive restarts.
- Choose Cache when the application can rebuild the value and accepts expiration or eviction.
- Choose Queues when work must be delivered to workers. For new integrations, start with Streams and consumer groups; use Pub/Sub only for real-time events that may be lost during a disconnect.
Create and operate
- Open Create project in the Console.
- Select the Valkey card for Key‑Value, Cache, or Queues.
- Select a plan, billing mode, and, where applicable, storage capacity.
- Wait for the service to become ready and copy the connection shown once.
- Use the project page, CLI, or API to inspect status, read the masked connection, and rotate credentials.
Do not provide an image, runtime, port, instance count, or IP rules: these fields are not part of the public managed Valkey configuration.
Choose by data lifecycle
Before creating an instance, decide whether the value can be rebuilt, must survive restarts, and has a responsible consumer that confirms each piece of work. This avoids using Cache for important data or turning a queue into a notification channel without retention. The catalog shows the current capacity and plan benefits; the project page shows status and connection details without revealing the previous password.
Reuse one connection or pool per process and close it during application shutdown. For workers, define time limits, retries with backoff, and a strategy for messages that remain pending. For Cache, record hits, misses, latency, and rebuild rate. For Key‑Value, monitor value sizes and key count. These metrics help choose the next plan without coupling the application to infrastructure details.
Next steps
Zenifra managed services
Learn about Zenifra managed services and choose ready-to-use data, caching, and asynchronous processing capabilities for your projects.
Connect to Valkey
Configure a secure TLS Valkey connection in Node.js, Python, Go, Java, PHP, .NET, Ruby, and Rust with the VALKEY_URL variable.