原文地址：https://blog.csdn.net/justry_deng/article/details/110299579
测试单机 redis 进行普通操作与 pipeline 操作
测试结果：
测试集群 redis 进行普通操作与 pipeline 操作 value
测试结果：
测试集群 redis 进行普通操作与 pipeline 操作 hash
测试结果：
核心依赖
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-data-redis</artifactId>
<exclusions>
<exclusion>
<groupId>io.lettuce</groupId>
<artifactId>lettuce-core</artifactId>
</exclusion>
</exclusions>
</dependency>
<dependency>
<groupId>redis.clients</groupId>
<artifactId>jedis</artifactId>
</dependency>
Pipeline 工具类
import com.niantou.redispipeline.author.JustryDeng;
import lombok.extern.slf4j.Slf4j;
import net.jcip.annotations.ThreadSafe;
import org.springframework.beans.BeansException;
import org.springframework.beans.factory.ObjectProvider;
import org.springframework.context.ApplicationContext;
import org.springframework.context.ApplicationContextAware;
import org.springframework.data.redis.connection.RedisClusterConnection;
import org.springframework.data.redis.connection.RedisCommands;
import org.springframework.data.redis.connection.RedisConnectionFactory;
import org.springframework.data.redis.connection.jedis.JedisClusterConnection;
import org.springframework.data.redis.core.RedisCallback;
import org.springframework.data.redis.core.RedisTemplate;
import org.springframework.data.redis.serializer.RedisSerializer;
import org.springframework.lang.NonNull;
import org.springframework.stereotype.Component;
import org.springframework.util.Assert;
import redis.clients.jedis.BinaryJedisCluster;
import redis.clients.jedis.Jedis;
import redis.clients.jedis.JedisCluster;
import redis.clients.jedis.JedisClusterConnectionHandler;
import redis.clients.jedis.JedisClusterInfoCache;
import redis.clients.jedis.JedisPool;
import redis.clients.jedis.Pipeline;
import redis.clients.jedis.Response;
import redis.clients.jedis.exceptions.JedisMovedDataException;
import redis.clients.jedis.exceptions.JedisNoReachableClusterNodeException;
import redis.clients.jedis.util.JedisClusterCRC16;
import java.lang.reflect.Field;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.Set;
import java.util.function.BiConsumer;
import java.util.function.BiFunction;
import java.util.function.Function;
import java.util.stream.Collectors;
/**
* redis pipeline 工具类RedisTemplate
*
* @author {@link JustryDeng}
* @since 2020/11/13 2:41:40
*/
@Slf4j
@Component
@ThreadSafe
@SuppressWarnings("unused")
public final class RedisPipelineUtil implements ApplicationContextAware {
private static RedisTemplate<?, ?> defaultRedisTemplate;
/**
* 获取key序列化器
*
* @return key序列化器
*/
@NonNull
public static RedisSerializer<Object> getKeySerializer() {
//noinspection unchecked
return (RedisSerializer<Object>) defaultRedisTemplate.getKeySerializer();
}
/**
* 获取value序列化器
*
* @return value序列化器
*/
@NonNull
public static RedisSerializer<Object> getValueSerializer() {
//noinspection unchecked
return (RedisSerializer<Object>) defaultRedisTemplate.getValueSerializer();
}
/**
* 获取hash-key序列化器
*
* @return hash-key序列化器
*/
@NonNull
public static RedisSerializer<Object> getHashKeySerializer() {
//noinspection unchecked
return (RedisSerializer<Object>)defaultRedisTemplate.getHashKeySerializer();
}
/**
* 获取hash-value序列化器
*
* @return hash-value序列化器
*/
@NonNull
public static RedisSerializer<Object> getHashValueSerializer() {
//noinspection unchecked
return (RedisSerializer<Object>)defaultRedisTemplate.getHashValueSerializer();
}
/**
* 流水线批量操作(单节点)
* <p>
* 注: SessionCallback是对RedisCallback的进一步封装， 不过我们都已经使用pipeline了， 那干脆直接用RedisCallback好了。
*
* @param biConsumer
* 批量操作逻辑
* @param paramList
* biConsumer用到的参数
* @return 结果集
*/
public static <R, P> List<R> pipeline4Standalone(BiConsumer<RedisCommands, P> biConsumer, final List<P> paramList) {
//noinspection unchecked
return (List<R>) defaultRedisTemplate.executePipelined((RedisCallback<R>) connection -> {
for (P p : paramList) {
biConsumer.accept(connection, p);
}
return null;
}, defaultRedisTemplate.getValueSerializer());
}
/**
* 流水线批量操作(单节点)
* <p>
* 注: SessionCallback是对RedisCallback的进一步封装， 不过我们都已经使用pipeline了， 那干脆直接用RedisCallback好了。
*
* @param redisTemplate
* 操作模板
* @param biConsumer
* 批量操作逻辑
* @param paramList
* biConsumer用到的参数集合
* @return 结果集
*/
public static <R, P> List<R> pipeline4Standalone(RedisTemplate<?, ?> redisTemplate, BiConsumer<RedisCommands, P> biConsumer, final List<P> paramList) {
//noinspection unchecked
return (List<R>) redisTemplate.executePipelined((RedisCallback<R>) connection -> {
for (P p : paramList) {
biConsumer.accept(connection, p);
}
return null;
}, redisTemplate.getValueSerializer());
}
/**
* 由于字符串使用的相对较多， 这里官(本)方(人)直接对字符串提供出来一个操作
* <p>
* @see RedisPipelineUtil#pipeline4ClusterSimpleStr(RedisTemplate, BiFunction, List)
*/
public static <R> List<R> pipeline4ClusterSimpleStr(BiFunction<Pipeline, PipelineParamSupplier<String>, Response<R>> biFunction,
List<String> paramList)
throws JedisMovedDataException {
return RedisPipelineUtil.pipeline4ClusterSimpleStr(defaultRedisTemplate, biFunction, paramList);
}
/**
* @see RedisPipelineUtil#pipeline4Cluster(JedisCluster, BiFunction, List)
*/
public static <R> List<R> pipeline4ClusterSimpleStr(@NonNull RedisTemplate<?, ?> redisTemplate,
BiFunction<Pipeline, PipelineParamSupplier<String>,
Response<R>> biFunction, List<String> paramList)
throws JedisMovedDataException {
RedisConnectionFactory redisConnectionFactory = redisTemplate.getConnectionFactory();
Assert.notNull(redisConnectionFactory, "redisConnectionFactory cannot be null");
RedisClusterConnection clusterConnection = redisConnectionFactory.getClusterConnection();
if (!(clusterConnection instanceof JedisClusterConnection)) {
throw new UnsupportedOperationException("cannot support RedisClusterConnection [" + clusterConnection.getClass().getName() + "]");
}
JedisCluster jedisCluster = ((JedisClusterConnection) clusterConnection).getNativeConnection();
@SuppressWarnings("unchecked") RedisSerializer<Object> keySerializer = (RedisSerializer<Object>)redisTemplate.getKeySerializer();
return RedisPipelineUtil.pipeline4ClusterSimpleStr(jedisCluster, keySerializer, biFunction, paramList);
}
/**
* 为保证keySerializer与jedisCluster是配套的，这里将此方法私有化，不对外提供
* <p>
* @see RedisPipelineUtil#pipeline4Cluster(JedisCluster, BiFunction, List)
*/
private static <R> List<R> pipeline4ClusterSimpleStr(@NonNull JedisCluster jedisCluster, RedisSerializer<Object> keySerializer,
BiFunction<Pipeline, PipelineParamSupplier<String>, Response<R>> biFunction,
List<String> paramList)
throws JedisMovedDataException {
List<StringSelfSupplier> supplierParamList = paramList.stream().map(x -> new StringSelfSupplier(x, keySerializer)).collect(Collectors.toList());
return RedisPipelineUtil.pipeline4Cluster(jedisCluster, biFunction, supplierParamList);
}
/**
* @see RedisPipelineUtil#pipeline4Cluster(RedisTemplate, BiFunction, List)
*/
public static <P extends PipelineParamSupplier<T>, T, R> List<R> pipeline4Cluster(BiFunction<Pipeline, PipelineParamSupplier<T>,
Response<R>> biFunction, List<P> paramList)
throws JedisMovedDataException {
return RedisPipelineUtil.pipeline4Cluster(defaultRedisTemplate, biFunction, paramList);
}
/**
* @see RedisPipelineUtil#pipeline4Cluster(JedisCluster, BiFunction, List)
*/
public static <P extends PipelineParamSupplier<T>, T, R> List<R> pipeline4Cluster(@NonNull RedisTemplate<?, ?> redisTemplate,
BiFunction<Pipeline, PipelineParamSupplier<T>, Response<R>> biFunction,
List<P> paramList)
throws JedisMovedDataException {
RedisConnectionFactory redisConnectionFactory = redisTemplate.getConnectionFactory();
Assert.notNull(redisConnectionFactory, "redisConnectionFactory cannot be null");
RedisClusterConnection clusterConnection = redisConnectionFactory.getClusterConnection();
if (!(clusterConnection instanceof JedisClusterConnection)) {
throw new UnsupportedOperationException("cannot support RedisClusterConnection [" + clusterConnection.getClass().getName() + "]");
}
return RedisPipelineUtil.pipeline4Cluster(((JedisClusterConnection) clusterConnection).getNativeConnection(), biFunction, paramList);
}
/**
* (使用JedisCluster，实现)流水线批量操作(集群)
*
* @param jedisCluster
* JedisCluster实例
* 其余参数
* @see JedisClusterPipeline#pipeline4Cluster(BiFunction, List)
*/
public static <P extends PipelineParamSupplier<T>, T, R> List<R> pipeline4Cluster(@NonNull JedisCluster jedisCluster,
BiFunction<Pipeline, PipelineParamSupplier<T>,
Response<R>> biFunction, List<P> paramList)
throws JedisMovedDataException {
JedisClusterPipeline jedisClusterPipeline = new JedisClusterPipeline(jedisCluster);
return jedisClusterPipeline.pipeline(biFunction, paramList);
}
@Override
@SuppressWarnings("rawtypes")
public void setApplicationContext(ApplicationContext applicationContext) throws BeansException {
ObjectProvider<PipelineDefaultRedisTemplateProvider> defaultRedisTemplateProvider = applicationContext.getBeanProvider(PipelineDefaultRedisTemplateProvider.class);
PipelineDefaultRedisTemplateProvider provider = defaultRedisTemplateProvider.getIfAvailable();
if (provider != null) {
RedisTemplate<?, ?> obtainedRedisTemplate = provider.obtainDefaultRedisTemplate();
RedisPipelineUtil.defaultRedisTemplate = obtainedRedisTemplate;
log.info(" use [{}] as the default RedisPipelineUtil's RedisTemplate", obtainedRedisTemplate);
} else {
// 初始化
Map<String, RedisTemplate> beansOfType = applicationContext.getBeansOfType(RedisTemplate.class);
Map.Entry<String, RedisTemplate> redisTemplateEntry = beansOfType.entrySet().stream()
.findFirst().orElseThrow(() -> new IllegalArgumentException(" cannot find any RedisTemplate"));
//noinspection
RedisPipelineUtil.defaultRedisTemplate = redisTemplateEntry.getValue();
log.info(" use [{}] as the default RedisPipelineUtil's RedisTemplate", redisTemplateEntry.getKey());
}
}
/**
* jedis使用pipeline操作redis-cluster辅助类
*
* 参考并整理自
* <a href="https://blog.csdn.net/youaremoon/article/details/51751991?utm_medium=distribute.pc_relevant.none-task-blog-searchFromBaidu-2.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-searchFromBaidu-2.control"/>
* <a href="https://www.cnblogs.com/xiaodf/p/11002184.html"/>
* <a href="https://blog.csdn.net/xiaoliu598906167/article/details/82218525?utm_medium=distribute.pc_aggpage_search_result.none-task-blog-2~all~first_rank_v2~rank_v25-10-82218525.nonecase&utm_term=pipeline%20redis%20%E8%BF%94%E5%9B%9E%E5%80%BC&spm=1000.2123.3001.4430"/>
*
* @author {@link JustryDeng}
* @since 2020/11/27 16:29:59
*/
static class JedisClusterPipeline {
private final JedisClusterConnectionHandler connectionHandler;
private final JedisClusterInfoCache infoCache;
public JedisClusterPipeline(JedisCluster jedisCluster) {
try {
Field connectionHandlerField = BinaryJedisCluster.class.getDeclaredField("connectionHandler");
boolean accessible = connectionHandlerField.isAccessible();
connectionHandlerField.setAccessible(true);
this.connectionHandler = (JedisClusterConnectionHandler) connectionHandlerField.get(jedisCluster);
connectionHandlerField.setAccessible(accessible);
Field cacheField = JedisClusterConnectionHandler.class.getDeclaredField("cache");
accessible = cacheField.isAccessible();
cacheField.setAccessible(true);
this.infoCache = (JedisClusterInfoCache) cacheField.get(connectionHandler);
cacheField.setAccessible(accessible);
} catch (Exception e) {
throw new RuntimeException(e);
}
}
/**
* (使用JedisCluster，实现)流水线批量操作(集群)
* <p>
* 注: 【据说】对集群redis进行pipeline, Jedis比Lettuce快。
* <p>
* 泛型说明
* <ul>
* <li>P：操作参数， 其需要实现{@link PipelineParamSupplier<T>}, 以获得 1.redis-key 2.最终进行pipeline的操作参数</li>
* <li>T：最终的pipeline操作参数</li>
* <li>R: 为返回的数据集合泛型</li>
* </ul>
*
* @param biFunction
* 批量操作逻辑
* @param paramList
* biFunction会用到的参数
* @throws JedisMovedDataException
* key对应的slot槽点变化时抛出
* @return 结果集
*/
<P extends PipelineParamSupplier<T>, T, R> List<R> pipeline(BiFunction<Pipeline, PipelineParamSupplier<T>, Response<R>> biFunction,
List<P> paramList) throws JedisMovedDataException {
// 从paramList中抽取到对应的redis-key集合
Map<byte[], P> redisKeyParamMap = paramList.stream().collect(Collectors.toMap(P::getRedisKey, Function.identity()));
Set<byte[]> allKeys = redisKeyParamMap.keySet();
Map<JedisPool, List<byte[]>> poolKeys = new HashMap<>(8);
// 刷新集群信息
connectionHandler.renewSlotCache();
for (byte[] key : allKeys) {
int slot = JedisClusterCRC16.getSlot(key);
JedisPool jedisPool = getJedisPoolFromSlot(slot);
if (poolKeys.containsKey(jedisPool)) {
List<byte[]> keys = poolKeys.get(jedisPool);
keys.add(key);
} else {
List<byte[]> keys = new ArrayList<>();
keys.add(key);
poolKeys.put(jedisPool, keys);
}
}
int size = allKeys.size();
List<R> result = new ArrayList<>(size);
List<Response<R>> responseList = new ArrayList<>(size);
poolKeys.forEach((JedisPool jedisPool, List<byte[]> keys) -> {
Jedis jedis = jedisPool.getResource();
Pipeline pipeline = jedis.pipelined();
try {
keys.forEach(key -> responseList.add(biFunction.apply(pipeline, redisKeyParamMap.get(key))));
} finally {
try {
pipeline.close();
} catch (Exception e) {
log.error(e.getMessage());
}
try {
jedis.close();
} catch (Exception e) {
log.error(e.getMessage());
}
}
});
responseList.forEach(response -> result.add(response.get()));
return result;
}
/**
* 根据槽点获取要对应使用的JedisPool
*/
private JedisPool getJedisPoolFromSlot(int slot) {
JedisPool connectionPool = infoCache.getSlotPool(slot);
if (connectionPool != null) {
// It can't guaranteed to get valid connection because of node assignment
return connectionPool;
} else {
// It's abnormal situation for cluster mode, that we have just nothing for slot, try to rediscover state
// 刷新集群信息
connectionHandler.renewSlotCache();
connectionPool = infoCache.getSlotPool(slot);
if (connectionPool != null) {
return connectionPool;
} else {
throw new JedisNoReachableClusterNodeException("No reachable node in cluster for slot " + slot);
}
}
}
}
/**
* 可指定{@link RedisPipelineUtil#defaultRedisTemplate}
* <p>
* 将会优先使用容器中被RedisTemplate4Pipeline标识了的RedisTemplate作为{@link RedisPipelineUtil#defaultRedisTemplate}，
* 如果没有，则以容器中第一个RedisTemplate作为{@link RedisPipelineUtil#defaultRedisTemplate}
*
* @author {@link JustryDeng}
* @since 2020/12/7 12:40:04
*/
public interface PipelineDefaultRedisTemplateProvider {
/**
* 指定{@link RedisPipelineUtil#defaultRedisTemplate}
*
* @return 要设置的{@link RedisPipelineUtil#defaultRedisTemplate}
*/
RedisTemplate<?, ?> obtainDefaultRedisTemplate();
}
/**
* Jedis操作redis-cluster时， pipeline操作参数提供器
*
* @author {@link JustryDeng}
* @since 2020/11/28 16:45:40
*/
public interface PipelineParamSupplier<T> {
/**
* 获取(序列化后的)redis key
* <p>
* P.S. 在使用Pipeline操作集群时，redis key使用这个方法获取。
* <p>
* 注: 这里之所以要将【获取redis key】抽取为一个方法，是因为相关逻辑中有多个地方会用到。 如果这些地方在将key对象序列化为byte[]时，
* 采用了不同的序列化方式， 那么可能存在数据槽slot定位不一致的问题， 进而(因代码不当)引起JedisMovedDataException异常。
* 为了避免上述问题，这里将获取redis-key的操作，抽取统一。
*
* @return redis key
*/
byte[] getRedisKey();
/**
* 获取pipeline操作需要的参数
* <p>
* P.S. 在使用Pipeline操作集群时，redis key使用这个方法获取。
* @return pipeline操作参数
*/
T getParam();
}
/**
* 官(本)方(人)对常用的字符串提供PipelineParamSupplier实现
*
* @author {@link JustryDeng}
* @since 2020/11/25 22:06:23
*/
public static class StringSelfSupplier implements PipelineParamSupplier<String> {
private final String str;
private final RedisSerializer<Object> keySerializer;
public StringSelfSupplier(String str, RedisSerializer<Object> keySerializer) {
this.str = str;
this.keySerializer = keySerializer;
}
@Override
public byte[] getRedisKey() {
return keySerializer.serialize(getParam());
}
@Override
public String getParam() {
return this.str;
}
}
}