Title: The Future of Modern Distributed SQL Databases: Data Replication
Time: June 21. The doors will open at 5:30 PM Pacific Daylight Time, and the talk will start at 6:00 PM.
Introduction:
Data replication is the most important topic in distributed database systems. To achieve high availability (HA) and fault tolerance, data are replicated several times. It’s not easy to keep the data from various replications consistent. In fact, the CAP theorem states that only two out of three desired properties—consistency, availability, and partition tolerance—can be achieved. In this talk, Henry will provide his insights on data replication.
Henry will also talk about Change Data Capture (CDC), a much-needed feature of modern databases. It captures the changes made to the data and then delivers those changes to the downstream systems. We design and implement a CDC system in TiDB called TiCDC. In this talk, Henry will elaborate on how to implement CDC in distributed databases, using TiDB as an example.
Henry Long
Head of Data Platform, PingCAP
Henry is head of Data Platform at PingCAP and joined PingCAP in 2018. Henry contributed a lot of open-source projects, and he was a TiDB maintainer and TiKV committer. He also launched TiUP. Henry is focusing on implementing the dataflow products in TiDB ecosystem to provide data migration, data replication, and backup restore capabilities.