HBase Redis 第三方… Spark 用户画像平台 现有的流程：  ES中定义标签的大宽表  通过Spark关联各种业务数据，插入到ES大 宽表。  高频查询的画像数据通过后台任务保存到加 速层：Hbase 戒者 Redis  实时标签通过Flink计算，然后写入Redis  用户画像平台可以从ES、Hbase、Redis查 询数据

(Zookeeper, etcd) › MapReduce (Hadoop, Spark) › Network File System (S3, HDFS) › Key-Value Storage (Redis, Aerospike) › Relational DBMS (PostgreSQL, MySQL) › NoSQL DBMS (MongoDB, Couchbase) https://github (Zookeeper, etcd) › MapReduce (Hadoop, Spark) › Network File System (S3, HDFS) › Key-Value Storage (Redis, Aerospike) › Relational DBMS (PostgreSQL, MySQL) › Coordination system (Zookeeper, etcd) › NoSQL Sources Simple › File on local FS › HTTP(S)-server External DBMS › MySQL › MongoDB › ClickHouse › Redis › ODBC Generic sources › Executable binary › Shared library 91 / 97 Query With Table SELECT OrderID

配置 max_bytes_before_external_sort max_bytes_before_external_group_by 2. 用户并发量一上来,负载太高 解决：目前是在中间加redis缓存

● Merge sharding tables ● SolidKey Typical Provider Config { Type: mysql, // mysql, mongodb, redis Listen: binlog, // binlog, kafka DataSource: user:pass@tcp(example.com:3306)/user, Table: user