About PingCAP • Since 2015 • Beijing, Shanghai, Guangzhou, Silicon Valley, Home • Open-Source infrastructure software: Codis, RebornDB • TiDB, TiKV, TiSpark Agenda • Design Goals • Architecture Evolution has the highest priority Tools matters • Automate Everything • Chaos tools • Debug tools • Infrastructure tools • Jira/Confluence How to tuning the system? 玄学调优 Metrics System Matters • 400+ metrics Releasing your product as an open source project will significantly improve it, especially for infrastructure software • Talent Acquire Linus‘s Law: “Given enough eyeballs, all bugs are shallow” ---

me © The Pythian Group Inc., 2019 �3 TiDB Created by PingCAP (Founded in April 2015 by 3 Infrastructure Engineers) © The Pythian Group Inc., 2019 ●SQL-based (MySQL[0] protocol). ●Distributed (ACID)

latency of report reduced from 30m to 2m • Reduced the cost of database servers. • Bigdata infrastructure adaption The challenges to storage module on HTAP scenario Isolation between OLTP and OLAP

About Me • Shen Li (申砾) • Tech Lead of TiDB, VP of Engineering • Netease / 360 / PingCAP • Infrastructure software engineer What is TiDB A Distributed, Consistent, Scalable, SQL Database that supports

物理机 ● 业务侵入性和性能损耗 ● 实时，秒级别有效性 行业现状 ● 起点：《Dapper, a Large-Scale Distributed Systems Tracing Infrastructure》 - Google 2010 ● OpenTracing ○ Zipkin ○ Uber Jaeger ● OpenCensus ● OpenTracing + OpenCensus

product option of TiDB, where users or organizations can deploy and manage TiDB on their own infrastructure with complete flexibility. With TiDB Self- Managed, you can enjoy the power of open source, distributed com/cn/blogs/security/get-the-full-benefits- �→ of-imdsv2-and-disable-imdsv1-across-your-aws-infrastructure/">AWS EC2 �→ now uses IMDSv2 as the default metadata service. TiDB supports �→ backing up analyzing the duration of Lock and Commit. 11.1.12 TiDB Best Practices on Public Cloud Public cloud infrastructure has become an increasingly popular choice for deploying and managing TiDB. However, deploying

product option of TiDB, where users or organizations can deploy and manage TiDB on their own infrastructure with complete flexibility. With TiDB Self- Managed, you can enjoy the power of open source, distributed com/cn/blogs/security/get-the-full-benefits- �→ of-imdsv2-and-disable-imdsv1-across-your-aws-infrastructure/">AWS EC2 �→ now uses IMDSv2 as the default metadata service. TiDB supports �→ backing up analyzing the duration of Lock and Commit. 11.1.12 TiDB Best Practices on Public Cloud Public cloud infrastructure has become an increasingly popular choice for deploying and managing TiDB. However, deploying

But this case is a rare occurrence. – Because the ISP dedicated network is used, the network infrastructure of this architecture has a high cost. – Five replicas are configured in three DCs in two cities (II) - Computing • TiDB Internal (III) - Scheduling 10.4.1.1 Preface Database is a generic infrastructure system. It is important to consider various user scenarios during the development process and should never operate a TiKV cluster with under 3 nodes in a production scenario. Depending on infrastructure, workload, and resiliency needs, you may wish to increase this number. It is worth noting that

But this case is a rare occurrence. – Because the ISP dedicated network is used, the network infrastructure of this architecture has a high cost. – Five replicas are configured in three DCs in two cities - Computing • TiDB Internal (III) - Scheduling 735 10.5.1.1 Preface Database is a generic infrastructure system. It is important to consider various user scenarios during the development process and should never operate a TiKV cluster with under 3 nodes in a production scenario. Depending on infrastructure, workload, and resiliency needs, you may wish to increase this number. It is worth noting that

But this case is a rare occurrence. – Because the ISP dedicated network is used, the network infrastructure of this architecture has a high cost. – Five replicas are configured in three DCs in two cities - Computing • TiDB Internal (III) - Scheduling 709 10.5.1.1 Preface Database is a generic infrastructure system. It is important to consider various user scenarios during the development process and should never operate a TiKV cluster with under 3 nodes in a production scenario. Depending on infrastructure, workload, and resiliency needs, you may wish to increase this number. It is worth noting that