ShardingSphere‐JDBC adopts decentralized architecture, applicable to high‐performance light‐weight OLTP application developed with Java; ShardingSphere‐Proxy provides static entry and all languages support increase the system capacity by removing database connection restrictions. It is also applicable to OLTP operation, which usually has sharding keys and can be routed to a single shard. So it is a wise choice circumstance, two requests waiting for each other will not happen, so there is no need for locking. Most OLTP operations use sharding keys to route to the only data node, which will make the system in a totally

损耗略高 损耗低 无中心化 是 否 是 静态入口 无 有 无 1.1.4 混合架构 ShardingSphere‐JDBC 采用 无中心化架构， 适用 于 Java 开发的高性能的轻量级 OLTP 应用； ShardingSphere‐Proxy 提供静态入口以及异构语言的支持，适用于 OLAP 应用以及对分片数据库 进行管理和运维的场景。 Apache ShardingSphere 以防止对一次请求对数据库连接占用过多所带来 的问题。该模式始终选择内存归并。 内存限制模式适用于 OLAP 操作，可以通过放宽对数据库连接的限制提升系统吞吐量；连接限制模式适 用于 OLTP 操作，OLTP 通常带有分片键，会路由到单一的分片，因此严格控制数据库连接，以保证在线 系统数据库资源能够被更多的应用所使用，是明智的选择。 自动化执行引擎 ShardingSphere 最初将使 并发。因此，ShardingSphere 在这里进行了 2 点优化： 1. 避免锁定一次性只需要获取 1 个数据库连接的操作。因为每次仅需要获取 1 个连接，则不会发生两 个请求相互等待的场景，无需锁定。对于大部分 OLTP 的操作，都是使用分片键路由至唯一的数据 节点，这会使得系统变为完全无锁的状态，进一步提升了并发效率。除了路由至单分片的情况，读 写分离也在此范畴之内。 2. 仅针对内存限制模式时才进行资源

ShardingSphere‐JDBC adopts a decentralized architecture, applicable to high‐performance light‐weight OLTP application developed with Java. ShardingSphere‐Proxy provides static entry and all languages support ShardingSphere‐JDBC adopts a decentralized architecture, applicable to high‐performance light‐weight OLTP application developed with Java. ShardingSphere‐Proxy provides static entry and all languages support brought more complex situations to the database. Apache ShardingSphere wants to solve massive data OLTP problem first and complete relevant OLAP support problem little by little. SQL SQL Supporting Status

ShardingSphere‐JDBC adopts a decentralized architecture, applicable to high‐performance light‐weight OLTP application developed with Java. ShardingSphere‐Proxy provides static entry and all languages support ShardingSphere‐JDBC adopts a decentralized architecture, applicable to high‐performance light‐weight OLTP application developed with Java. ShardingSphere‐Proxy provides static entry and all languages support brought more complex situations to the database. Apache ShardingSphere wants to solve massive data OLTP problem first and complete relevant OLAP support problem little by little. 4.3. Sharding 37 Apache

platform Reporting Databases App BI Console Ad hoc Data Platform - What You Think It Is OLTP DBs App Data Warehouse / Data Lake ETL Analytical DBs Reporting BI Console Ad hoc Data Platform point / short access ○ Row format for OLTP ○ Columnar format for OLAP ● Workload Interference ○ A single large analytical query might cause disaster for your OLTP workload Fundamental Conflicts ● Use Use different types of databases ○ For live and fast data, use an OLTP specialized database or NoSQL ○ For historical data, use Hadoop / analytical database ● Offload data via the ETL process into your

Syntax (MySQL 互換) Distributed Transactions (分散トランザクション) Cloud Native (クラウドネイティブ志向) Minimize ETL (OLTP と OLAP のサポート) High Availability (高可用性) Open Source Conference 2022 Online/Spring 12 TiDB の特徴 Syntax (MySQL 互換) Distributed Transactions (分散トランザクション) Cloud Native (クラウドネイティブ志向) Minimize ETL (OLTP と OLAP のサポート) High Availability (高可用性) Open Source Conference 2022 Online/Spring 13 TiDB の特徴 (Horizontal Syntax (MySQL 互換) Distributed Transactions (分散トランザクション) Cloud Native (クラウドネイティブ志向) Minimize ETL (OLTP と OLAP のサポート) High Availability (高可用性) Open Source Conference 2022 Online/Spring 16 TiDB の特徴 (MySQL

methods reported on the following pages. About the Benchmark: TPC‐C and tpcc‐mysql TPC‐C is an OLTP industry standard benchmark developed and maintained by the Transaction Processing Performance Council processing orders. It is a mixture of read‐only and update‐intensive transactions that represent complex OLTP application activities. The benchmark implements five types of transactions (new order, payment, box” code, downloaded from the Percona GitHub site. 2. Hardware Configurations In the past, OLTP systems were frequently bottlenecked by IO, meaning CPUs were constantly waiting on HDDs to respond

81PB存储（增长 中） 4. 2.5PB 或 25PB 原始数据 按10x压缩率 1 2 3 4 Greenplum在摩根士丹利 OLTP - Online transaction processing - 联机事务处理 出色的OLTP特性 天生的优势 ● 行式存储 ● 索引 ● 直接分发 ● 完整的增删改 Greenplum 6 增强 ● 并发修改、删除 ● 系统性的优化事务和锁 系统性的优化事务和锁 26 Pivotal Confidential–Internal Use Only 行式存储 表‘SALES’ 表‘SALES’ ■ 更适合OLTP负载 ■ 高效更改和删除 ■ 适合需要全部或者多数列的查询 表 orders 索引 Greenplum支持以下索引: • Btree • Bitmap • Gist • GIN • BRIN (开发中) - 混合事务/分析处理 Gartner技术成熟度曲线 OLTP-OLAP独立部署 OLTP数据库 OLAP数据仓库 ■ 实时性 ■ 数据同步复杂性 ■ 应用复杂性 HTAP HTAP = ? ■ 卓越的OLAP特性 ■ 出色的OLTP特性 ■ 多态存储 ■ 有效的并发和资源管理 OLTP-OLAP独立部署 OLTP数据库 OLAP数据仓库 ■ 实时性 ■ 数据同步复杂性

动态扩展 分布式事务 HTAP 真正⾦融级⾼可⽤ 适⽤场景 适⽤场景 对数据⼀致性及⾼可靠、系统⾼可⽤、可扩展性、容灾要求较⾼的⾦融⾏业属性的场景 对存储容量、可扩展性、并发要求较⾼的海量数据及⾼并发的 OLTP 场景 Real-time HTAP 场景 数据汇聚、⼆次加⼯处理的场景 真正⾦融级⾼可⽤ UCloud 云上 云上 TiDB 架构⽰意图 架构⽰意图 TiDB TiDB Serverless ⽬录 failover)，⽆需⼈⼯介⼊。 产品优势 分布式NewSQL数据库 TiDB Copyright © 2012-2021 UCloud 优刻得 13/120 适⽤场景 适⽤场景 为⽤⼾提供⼀站式 OLTP (Online Transactional Processing)、OLAP (Online Analytical Processing)、HTAP 解决⽅案。TiDB 适合⾼可⽤、强⼀致要求较⾼、数据规模较⼤等各种应⽤ 现故障时系统可⾃动进⾏切换，确保系统的 RTO <= 30s 及 RPO = 0。 对存储容量、可扩展性、并发要求较⾼的海量数据及⾼并发的 对存储容量、可扩展性、并发要求较⾼的海量数据及⾼并发的 OLTP 场景 场景 随着业务的⾼速发展，数据呈现爆炸性的增⻓，传统的单机数据库⽆法满⾜因数据爆炸性的增⻓对数据库的容量要求，可⾏⽅案是采⽤分库分表的中间件产品或者 NewSQL 数据库替代、 采⽤⾼端的存储设备等，其中性价⽐最⼤的是