Despite the importance of cloud databases as a core foundation for applications, many businesses face challenges in identifying database performance issues. Visibility into database performance is difficult due to a wide range of incomplete tools that can be difficult to install, configure, and maintain. While these tools may provide a wide range of statistics, they lack a standard methodology for analyzing the statistics to identify performance problems. In this session, learn how Amazon Relational Database Service (Amazon RDS) changes this by providing database performance monitoring that is automatically configured, easy to use, and based on a clear actionable methodology.

2. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Using Performance Insights to
Optimize Database Performance
Kyle Hailey
Principal Product Manager
Amazon RDS
D A T 4 0 2

3. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Agenda
What is Performance Insights?
Sampling
Average active sessions (AAS)
Bottleneck analysis
Exploring Performance Insights

4. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

5. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
What is Amazon RDS Performance Insights?
Customers asked for
• Visibility into performance of Amazon Relational Database Service
(Amazon RDS) databases
• Want to optimize cloud database workloads
• Easy tool
• Often only part-time DBA or no DBA
• Single pane of glass

6. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
• Released 2016
• OS metrics
• Process/Thread list
• Up to 1 second granularity
First step: Amazon RDS Enhanced Monitoring

7. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Introducing: Performance Insights
• Dashboard
• DB load
• Adjustable timeframe
• Filterable by attribute (SQL, User, Host, Wait)
• SQL causing load
• Phased Amazon RDS delivery
• Amazon Aurora, Amazon RDS for MySQL, PostgreSQL, Oracle, SQL Server, MariaDB
• Guided discovery of performance problems
• For both beginners & experts
• Core metric “database load”

8. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
• All engines have a connections list showing
• Active
• Idle
• We sample every second
• For each active session, collect
• SQL
• State: CPU, I/O, lock, commit log wait, and more
• Host
• User
• Expose as “average active sessions” (AAS)
What is “database load”?

9. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Performance Insights dashboard

10. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

11. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Sampling is like film

12. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Query run often
Fast query run rarely
Slow query
User 1
User 2
User 3
Time
Sampling every second

13. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session state
idleidle idle idleQuery 1 Query 2 Query 3
Time

14. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS load graph
User 1
User 2
User 3
User 4

15. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS load graph
User 1
User 2
User 3
User 4
Active sessions
=
1
2
3
4

16. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session
Time
User 1
1 2 3 4 5 6 7 8 9 10
Active sessions
1
2
3
4

17. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session
Time
User 1
User 2
1 2 3 4 5 6 7 8 9 10
Active sessions
1
2
3
4

18. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session
Time
User 1
User 2
1 2 3 4 5 6 7 8 9 10
Active sessions
1
2
3
4

19. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Average active session
Time
User 1
User 2
1 2 3 4 5 6 7 8 9 10
Active sessions
1
2
3
4

20. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Average active session
Time
User 1
User 2
User 3
User 4
User 5
User 6
User 7
User 8
User 9
User 10
1 2 3 4 5 6 7 8 9 10
Active sessions
1
2
3
4

21. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session state
idleidle idle idleQuery 1 Query 2 Query 3
Time

22. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Active session state
Time
CPU IO Wait
idleidle idle idleQuery 1 Query 2 Query 3

23. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS by session state

24. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Showing per second samples

25. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS over one minute averages

26. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS compared to max CPU
Max vCPU

27. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

28. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AAS rules: Using CPU count as yardstick
 AAS < 1
Database is not blocked
 AAS ~= 0
Database basically idle
Problems are in the APP not DB
 AAS < # of CPUs
CPU available
Are any single sessions 100% active?
• AAS > # of CPUs
Could have performance problems
 AAS >> # of CPUS
There is a bottleneck

29. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
When users say the database is slow …
You prove that it’s not the database
AAS = 0

30. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Also useful for sizing
• If CPU load significantly less than #vCPU then oversized
• If CPU load Is > #vCPU undersized

31. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

32. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Accessing Performance Insights

33. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Access to Performance Insights

34. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

35. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Max vCPU
Max
vCPU

36. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
CPU bottleneck
Bottleneck

37. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Customer use case: CPU bottleneck
Wait
States

38. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
CPU bottleneck
CPU

39. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
CPU bottleneck
CPU
Load
SQL
with high CPU

40. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Click and drag

41. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Zoom in

42. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

43. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Wait bottleneck

44. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Wait bottleneck

45. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

46. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Other dimensions

47. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Top host by SQL statement

48. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

49. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Performance Insights across DB engines
• Performance Insights supports
• Amazon Aurora
• MySQL
• Postgres
• Amazon RDS
• MySQL
• Postgres
• Oracle
• RDS SQL Server and MariaDB forthcoming
• Interface is the same across different engines
• Allows DBA to do performance work across different engines easily
• Dashboard content same
• Only difference is the wait event names, which are engine dependent

50. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

51. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Amazon Aurora MySQL—Five users

52. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Amazon Aurora PostgreSQL—Five users

53. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Oracle—Five users

54. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
PostgreSQL —Five users

55. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MySQL—Five users

56. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
What is available?
• Available
• Engines
• Amazon Aurora PostgreSQL
• Amazon Aurora MySQL 5.6 1.17.3 and higher
• Amazon RDS for PostgreSQL 10
• Amazon RDS for MySQL 5.6.41+ and 5.7.22+
• Amazon RDS for Oracle
• Functionality
• DB load chart
• Top N table
• Wait, user, host, SQL
• API/SDK
• Long-term data retention
• Alerts through Amazon CloudWatch

57. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
What is Coming
• Available
• Engines
• Amazon Aurora PostgreSQL
• Amazon Aurora MySQL 5.6 1.17.3 and higher
• Amazon RDS for PostgreSQL 10
• Amazon RDS for MySQL 5.6.41+ and 5.7.22+
• Amazon RDS for Oracle
• Functionality
• DB load chart
• Top N table
• Wait, user, host, SQL
• API/SDK
• Long-term data retention
• Alerts through Amazon CloudWatch
• Coming
• Engines
• Amazon Aurora MySQL 5.7
• Amazon RDS for MariaDB
• Amazon RDS for SQL server
• Functionality
• SQL execution plan
• SQL stats
• OS and DB statistics

58. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Summary: Amazon RDS Performance Insights
 DB load: Average active sessions
 Identifies database bottlenecks
 Easy
 Powerful
 Top SQL
 Identifies source of bottleneck
 Enables problem discovery
 Adjustable time frame
 Hour, day, week, and longer
 Questions:
 rdspi@amazon.com

59. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

60. “With Performance Insights, we have
been able to quickly isolate slow and
underperforming queries.”
Matt Finlayson
Vice President of Engineering
Cloudability

61. “Not only has [Performance Insights]
saved us a lot of time in diagnosing
queries, it has also helped us lower our
costs. ”
Steve Atherton
CTO
Catalyz

62. “One of the most valuable tools AWS
provides for Aurora is Performance
Insights. Our DBA uses this dashboard
on a daily basis”
Stephen Sciarini
IT Manager
New Innovations

63. Before “our diagnostics process was
laborious … With Performance Insights,
we open one tool and quickly and easily
see where we have problems.”
Grant Evans
DBA & AWS Administrator
Enova

64. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark
Alexa + Performance Insights
New integration between Alexa & Performance
Insights can be your assistant DBA
Uses the Performance Insights API to identify
bottlenecks in Amazon RDS
Get actionable suggestions such as on-demand
scaling, DBA notifications & paging
Go to the Slalom booth (#1438) to see a live demo
and learn more!

65. Thank you!
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Kyle Hailey
Principal Product Manager
Amazon RDS

66. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.