Webinar Slides: MySQL Native Replication vs. Tungsten Clustering
0 likes139 views
The document compares MySQL native replication and Tungsten replicator, highlighting their features, limitations, and different topologies. While native replication is easy to set up and free, it is limited to a one-to-many topology with basic filtering and lacks automatic failover. Tungsten replicator, on the other hand, offers complex filtering, heterogeneous replication, and is suitable for a variety of databases and topologies, ensuring high availability and disaster recovery options.
Webinar Slides: MySQL Native Replication vs. Tungsten Clustering
- 1. MySQL Native Replication vs Tungsten 1
- 2. Topics In this short course we will: • Features and limitations of Native MySQL Replication • Explore Tungsten Replicator • Review various Replicator Topologies • Talk about Replication Filtering • Learn about Tungsten Cluster 2 2
- 4. MySQL Native Replication 4 4 • It’s easy to setup and it’s free • Embedded within MySQL Engine • One-to-Many Topology only • Parallel Threads by Schema • Stale reads possible on slave • No easy mechanism to failover • Manual operation required to switch roles • Only Basic Include/Exclude filtering available • Many additional components required to function as a fully fault tolerant cluster • MySQL to MySQL Only
- 6. Tungsten Replicator 6 6 • Open Source (free!!) and Commercial Builds available • One-to-Many / Many-to-One Topologies • Parallel Threads by Schema • Complex filtering • Heterogeneous replication – Extract from MySQL (Incl RDS and Google SQL) or Oracle – Apply to MySQL (Incl RDS and Google SQL), Oracle, Redshift, Vertica, Hadoop, Mongo, Elasticsearch, Kafka (and more coming soon)
- 7. Tungsten Replicator – How it works 7 7 DBMS Logs Download transactions via network Apply using JDBC THL = Events + Metadata MySQL MySQL Binary Logging Option 1: Local Install Extractor reads directly from the logs, even when the DBMS service is down. This is the default. Option 2: Remote Extractor gets log data via MySQL Replication Slave protocols (which requires the DBMS service to be online). This is how we handle RDS extraction tasks. Extractor Options Master Replicator: Extractor THL 2 1 Slave Replicator: Applier THL MySQL
- 8. Tungsten Replicator - Topologies 8 8
- 9. Tungsten Replicator - Topologies 9 9
- 10. Tungsten Replicator – Complex Heterogeneous Topology 10 10
- 11. Filtering with Tungsten Replicator
- 12. Types of Filters 1212 • Auditing – Can be used for tracking database updates • Content – Modify contents of transactions, such as enumerating integers or remove columns, tables, or schemas • Logging – Log information about transactions in replicator log file • Optimization – Optimize statements and improve speed of updates on destination • Transformation – Rename, reformat schemas and tables, even merge schemas into a single schema • Validation – Validation and Consistency checking of data • Miscellaneous – Others • Over 40 filters included with Tungsten Replicator which can be ordered as desired!
- 13. Popular Filters 1313 • DatabaseTransform filter – rename databases or tables, supports regex • InsertOnly filter – filter to include only “INSERT” events, ignoring others • ReplicateColumn filter – replicate columns, either by inclusion or by exclusion • Replicate filter - explicit inclusion or exclusion of schemas and tables to replicate. Can use wildcards • SkipEventsByType filter – skip events for entire schemas or tables. For instance, skip all deletes to SALES.INVOICE • TimeDelay filter – delays writing events to THL. Allows having a slave that is behind the master for a specified time period. • SkipRowByData – replicate rows matching a condition
- 14. Tungsten Replicator - Filtering 14 14 Master replicator THL Parallel Queue (Events+ metadata) Slave Extract Filter Apply Extract Filter Apply Extract Filter Apply Extract Filter Apply Extract Filter Apply StageStage Stage Slave Replicator Pipeline remote-to-thl thl-to-q q-to-dbms
- 15. Tungsten Cluster for HA/DR
- 16. Tungsten Cluster 16 16 • Sits outside the MySQL Engine • Uses Commercial release of Tungsten Replicator at its core • Connector layer provides transparent read/write splitting • Automatic Failover • Simple, single command to switch masters • Tungsten Replicator can be used to provide cluster-slaves • Add/Remove nodes to scale reads • Zero Downtime Maintenance • Works with all “flavours” of MySQL (Community, Oracle, Percona and Maria) • Multi-Master and Composite Topologies for expanding HA/DR capabilities
- 18. Tungsten Multi-Site/Multi-Master Cluster Architecture 18 18
- 19. Tungsten Cluster using Tungsten Replicator to Apply to a Cluster Slave 19 19
- 20. Next Steps • If you are interested in knowing more and would like to try it out for yourself, please contact our sales team who will be able to take you through the details and setup a POC – sales@continuent.com • Read the documentation at http://docs.continuent.com/ • Follow us on Social Media – Facebook / Twitter / LinkedIn – Tungsten University YouTube channel! http://tinyurl.com/TungstenUni • Visit the events calendar on our website for upcoming Webinars and Training Sessions https://www.continuent.com/events/ 20 20
- 21. For more information, contact us: MC Brown VP Products mc.brown@continuent.com Chris Parker Director, Professional Services EMEA & APAC chris.parker@continuent.com Matthew Lang Director, Professional Services Americas matthew.lang@continuent.com Eero Teerikorpi Founder, CEO eero.teerikorpi@continuent.com +1 (408) 431-3305 Eric Stone COO eric.stone@continuent.com