Analyse_Amazon_Aurora_Workloads_with_Performance_Insights.pdf

Presentation from my talk given at AWS Sofia Community Day 2019

Database performances are often bottleneck in most of the workloads. Amazon RDS made life easier but still, most of the users have a problem how to find the root causes of database performance problems. In this session, you will have a chance to learn and see how you can use the Performance Insights dashboard and learn to analyse certain performance issues. Performance Insights makes detecting performance problems easy for both experts and non-experts. Besides that, you will be able to get a deep look and learn about Amazon Aurora performances.

Transcript

1. Analyse Amazon Aurora Workloads with Performance Insights AWS Community Day Sofia Dzenan Dzevlan TN-TECH @sqlheisenberg
2. whoami • AWS Consultant • Cloud and DevOps expert at TN-TECH • AWS Bosnia User Group leader & co-organizer • DevOps / Database / Cloud enthusiast @sqlheisenberg
3. What is Amazon Aurora • Database designed for the cloud • Simplicity and cost-effectiveness of open source databases • Simplicity and cost-effectiveness of open source databases • Pay as you go pricing • Delivered as a manged service
4. Re-Imagining the relational databases • Scale-out, distributed architecture • Service-oriented architecture leveraging AWS services • Automate administrative tasks – fully managed service •
5. Scale-out, distributed architecture Purpose-built log-structured distributed storage system designed for databases Storage volume is striped across hundreds of storage nodes distributed over 3 different availability zones Six copies of data, two copies in each availability zone to protect against AZ+1 failures
6. Leveraging AWS Services • Invoke lambda functions from stp/triggers Load data from S3, store backups on S3 • IAM authentication Upload systems metrics and audit logs to CloudWatch
7. Automate administrative tasks
8. Why to move to Amazon Aurora • Higher performances, up to 5% • Better availability and durability • Reduces costs • Easy migration, no application changes
9. Aurora Performances • 5x faster than MySQL; ??? • 3x faster than PostgreSQL; ???
10. Aurora vs MySQL
11. Aurora vs PostgreSQL
12. How did Aurora achieve this
13. IO traffic in Aurora database
15. Asynchronous group commits
16. IO traffic in Aurora (storage node)
17. Thread pooling
18. Performances insights • Dashboard showing load on DB • Identifies source of bottlenecks • Adjustable time frame
19. Demo
20. “Performances only matters if your database is up”
21. • Q&A • You forgot to ask, contact me on: dzenan.dzevlan@gmail.com
22. Thank you !