Boosting Redis Performance: Deep Dive into Lettuce and Redisson Pipeline Support in Spring Boot

Introduction

Redis uses a client‑server TCP model with request/response semantics. When many commands need to be executed in batch, Redis offers the Pipeline feature to reduce round‑trip time (RTT) and improve throughput. This article examines how the Lettuce and Redisson clients in a Spring Boot environment support pipeline, provides performance measurements, and shares lessons learned from real‑world usage.

Pipeline Fundamentals

In a normal request, the client sends a command, blocks waiting for the server’s response, and then processes the result. The three‑step flow is:

Client issues a request and blocks on the socket.

Server processes the command and returns a result.

Client receives the result and resumes execution.

Pipeline reduces the number of network round‑trips by sending multiple commands in one write operation and receiving all replies in a single read, thus cutting down both RTT and the number of costly user‑to‑kernel context switches.

Basic Usage and Performance Comparison

We benchmarked inserting 100,000 SET entries using three approaches:

@Slf4j
public class RedisPipelineTestDemo {
    public static void main(String[] args) {
        // Jedis single inserts
        Jedis jedis = new Jedis("10.101.17.180", 6379);
        String zSetKey = "Pipeline-test-set";
        int size = 100000;
        long begin = System.currentTimeMillis();
        for (int i = 0; i < size; i++) {
            jedis.sadd(zSetKey + i, "aaa");
        }
        log.info("Jedis single insert time: {}ms", System.currentTimeMillis() - begin);

        // Jedis pipeline
        Pipeline pipeline = jedis.pipelined();
        begin = System.currentTimeMillis();
        for (int i = 0; i < size; i++) {
            pipeline.sadd(zSetKey + i, "bbb");
        }
        pipeline.sync();
        log.info("Jedis pipeline time: {}ms", System.currentTimeMillis() - begin);

        // Redisson pipeline (RBatch)
        Config config = new Config();
        config.useSingleServer().setAddress("redis://10.101.17.180:6379");
        RedissonClient redisson = Redisson.create(config);
        RBatch batch = redisson.createBatch();
        begin = System.currentTimeMillis();
        for (int i = 0; i < size; i++) {
            batch.getSet(zSetKey + i).addAsync("ccc");
        }
        batch.execute();
        log.info("Redisson pipeline time: {}ms", System.currentTimeMillis() - begin);

        // Cleanup
        pipeline.close();
        jedis.close();
        redisson.shutdown();
    }
}

Results:

Jedis single insert: 162,655 ms

Jedis pipeline: 504 ms

Redisson pipeline: 1,399 ms

Pipeline dramatically outperforms individual commands, with Jedis pipeline being the fastest in this test.

Real‑World Application

In a production scenario, we needed to update icon styles for millions of users during holidays. The workflow stores per‑user icon identifiers in Redis, and we batch‑write these values using pipeline to avoid excessive latency.