Best Practices for Using Alibaba Cloud Redis: Key Design, Command Usage, Client Configuration, and Tools

Introduction

This guide presents development standards for using Alibaba Cloud Redis, explaining key design, command usage, client usage, and related tools to reduce issues during Redis usage.

1. Key Design

Key Naming

Recommendation: readability and manageability – use business or database name as a prefix and separate parts with colons, e.g., business:table:id .

Example: ugc:video:1

Recommendation: keep keys concise; avoid special characters, spaces, line breaks, or quotes.

Example of shortened key: user:{uid}:friends:messages:{mid} → u:{uid}:fr:m:{mid}

Value Design

Enforce: reject big keys to prevent network traffic spikes and slow queries.

String values should be ≤10KB; hash, list, set, and sorted‑set elements should not exceed 5,000. Example of a bad key: a list with 2 million elements.

For non‑string big keys, avoid DEL ; use HSCAN , SSCAN , or ZSCAN for incremental deletion and be aware of expiration side‑effects.

Recommendation: choose appropriate Redis data types and configure memory encoding (e.g., ziplist) while balancing memory savings and performance.

Example commands:

set user:1:name tom

set user:1:age 19

set user:1:favor football

Correct bulk insertion:

hmset user:1 name tom age 19 favor football

Key Lifecycle

Set expiration with EXPIRE ; if possible, stagger expirations to avoid spikes. Monitor idle time for keys that do not expire.

2. Command Usage

O(N) Commands

Avoid commands that scan large N without limits; use HSCAN , SSCAN , ZSCAN instead of HGETALL , LRANGE , etc., when N is unknown.

Disable Dangerous Commands

Prohibit KEYS , FLUSHALL , FLUSHDB in production; rename or block them, or replace with progressive SCAN processing.

Select Usage

Redis multiple databases are weakly supported; avoid relying on them for isolation.

Batch Operations

Use pipelines for bulk operations (e.g., MGET , MSET ) but limit batch size (≈500 commands) and be aware of atomicity differences.

Differences:

Native commands are atomic; pipelines are not.

Pipelines can bundle different commands.

Both client and server must support pipelines.

Transactions

Redis transactions are limited (no rollback) and, in cluster mode, all keys in a transaction must reside in the same slot (use hashtags to co‑locate keys).

Lua Scripts in Cluster

All keys must be passed via the KEYS array; otherwise the script fails.

All keys must belong to the same slot.

Monitor Command

Use MONITOR sparingly; avoid long‑running sessions.

3. Client Usage

Do not share a single Redis instance across unrelated applications; separate business domains and expose shared data via services.

Use connection‑pooled clients to control connections and improve efficiency.

In high‑concurrency scenarios, add circuit‑breaker mechanisms (e.g., Netflix Hystrix).

Secure access with strong passwords and enable SSL when needed (Alibaba Cloud Redis supports SSL).

Configure appropriate maxmemory‑policy based on workload; default is volatile‑lru . Other policies include:

allkeys‑lru

allkeys‑random

volatile‑random

volatile‑ttl

noeviction

4. Related Tools

Data synchronization: redis‑port

Big‑key search tools

Hot‑key detection (uses MONITOR , use briefly)

5. Appendix – Deleting Big Keys

Use pipelines to accelerate deletion.

Hash: HSCAN + HDEL

List: LTRIM

Set: SSCAN + SREM

Sorted Set: ZSCAN + ZREM

These methods allow gradual removal of large data structures without blocking the Redis server.