Cassandra o que nao te contaram
Download as PPTX, PDF0 likes53 views
Rafael Monteiro discusses common mistakes made when using Cassandra as an infrastructure for databases. Some key mistakes include sharing storage across nodes, putting the commit log and data on the same disks, and putting a load balancer between the client and cluster. Other design mistakes include using Cassandra as a relational database when the data is not denormalized, having too many materialized views, and filtering by columns that are not in the primary key. Proper data modeling and running nodetool repair can help avoid issues. The presentation encourages using the right tool for the job.
Cassandra o que nao te contaram
- 1. Cassandra o que não te contaram Rafael Monteiro e Pereira Software Engineer
- 2. About me • Rafael Monteiro • Mestrando em Inteligência Artificial • Software Engineer
- 5. Lets talk about ... • Cassandra
- 6. Common infrastructure mistakes • Shared Storage • SAN • NAS
- 7. Common infrastructure mistakes • Put Commit Log and Data on the same Disks
- 8. Common infrastructure mistakes • Put a Load Balancer between the cliente and the cluster
- 9. Common infrastructure mistakes • Huge JVM HEAPS! • 4-8GB = Low • 8-12GB = Ok • 12-16GB = Careful • >16 GB = Bad Idea
- 11. Design Mistakes • If your data is relational – USE a RDBMS
- 12. Design Mistakes • Use Cassandra as a Key – Value Store • KEY - PAYLOAD
- 13. Design Mistakes • Too many materialized views
- 14. Design Mistakes • Filter by a column that is not on the key.
- 15. Conclusion • Model it RIGHT and you will be fine! • Don’t forget running nodetool repair ! • Use the right tool for the job
- 16. Obrigado!!!! Feedback ? Contato ? E-mail: [email protected] Github: github.com/rafamonteiro Twitter: @monteiro_io LinkedIn: lnked.in/rafamonteiro