Part #1 Part #2 Part #3. Background Engineering Oracle Rant

Transcription

1 @andy_pavlo

2 Part #1 Part #2 Part #3 Background Engineering Oracle Rant

3 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES Admin SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' s Self-Adaptive Databases

4 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Adaptive Databases

5 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Adaptive Databases

6 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Adaptive Databases

7 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Adaptive Databases

8 AUTONOMOUS DBMSs 3 SELF-ADAPTIVE DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Adaptive Databases Index Selection Partitioning / Sharding Data Placement

9 AUTONOMOUS DBMSs 4 SELF-TUNING DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME s Self-Tuning Databases Index Selection Partitioning / Sharding Data Placement

10 AUTONOMOUS DBMSs 4 SELF-TUNING DATABASES SELECT * FROM A JOIN B ON A.ID = B.ID WHERE A.VAL > 123 AND B.NAME LIKE 'XY%' s Self-Tuning Databases Admin Tuning Algorithm A.ID A.VAL B.ID B.NAME AutoAdmin Optimizer Cost Model

11 AUTONOMOUS DBMSs 4 SELF-TUNING DATABASES Number of Knobs s Self-Tuning Databases Knob Configuration

12 AUTONOMOUS DBMSs 5 CLOUD MANAGED DATABASES 2010s Cloud Databases

13 AUTONOMOUS DBMSs 5 CLOUD MANAGED DATABASES Initial Placement Tenant Migration 2010s Cloud Databases

14 Why is this previous work insufficient?

15 AUTONOMOUS DBMSs 7 A BRIEF HISTORY Problem #1 Human Judgements Problem #2 Reactionary Measures

16 What is different this time?

17 AUTONOMOUS DATABASES WHY NOW? Better hardware. Better machine learning tools. Better appreciation for data. We seek to complete the circle in autonomous databases.

18 CARNEGIE MELLON UNIVERSITY 10 RESEARCH PROJECTS OtterTune Existing Systems Peloton New System

19 OtterTune ottertune.cs.cmu.edu Database Tuning-as-a-Service Automatically generate DBMS knob configurations. Reuse data from previous tuning sessions. Supported Systems

20 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TARGET DATABASE INSTALL AGENT

21 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

22 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

23 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

24 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

25 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

26 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

27 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

28 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

29 OTTERTUNE 12 AUTOMATIC DBMS TUNING SERVICE CONTROLLER COLLECTOR TUNING MANAGER Internal Repository TARGET DATABASE INSTALL AGENT Configuration Recommender Metric Analyzer Knob Analyzer

30 OTTERTUNE TPC-C TUNING Default Scripts RDS DBA Throughput (txn/sec) OtterTune AUTOMATIC DATABASE MANAGEMENT SYSTEM TUNING THROUGH LARGE-SCALE MACHINE LEARNING SIGMOD 2017

31 Peloton pelotondb.io Self-Driving Database System In-memory DBMS with integrated ML/RL framework. Designed for autonomous operations.

32 PELOTON 15 THE SELF-DRIVING DBMS WORKLOAD HISTORY TARGET DATABASE

33 PELOTON 15 THE SELF-DRIVING DBMS WORKLOAD HISTORY TARGET DATABASE FORECAST MODELS

34 PELOTON 15 THE SELF-DRIVING DBMS "THE BRAIN" WORKLOAD HISTORY Search Tree ACTION CATALOG TARGET DATABASE FORECAST MODELS

35 PELOTON 15 THE SELF-DRIVING DBMS "THE BRAIN" WORKLOAD HISTORY Search Tree TARGET DATABASE FORECAST MODELS ACTION CATALOG ACTION SEQUENCE

36 PELOTON 15 THE SELF-DRIVING DBMS "THE BRAIN" WORKLOAD HISTORY Search Tree TARGET DATABASE FORECAST MODELS ACTION CATALOG ACTION SEQUENCE

37 PELOTON 15 THE SELF-DRIVING DBMS "THE BRAIN" WORKLOAD HISTORY Search Tree ACTION CATALOG??? TARGET DATABASE FORECAST MODELS ACTION SEQUENCE

38 Queries Per Hour PELOTON BUS TRACKING APP WITH ONE-HOUR HORIZON Actual Ensemble (LR+RNN) Predicted Jan 11-Jan 13-Jan 15-Jan 17-Jan QUERY-BASED WORKLOAD FORECASTING FOR SELF-DRIVING DATABASE MANAGEMENT SYSTEM SIGMOD 2018

39 Queries Per Hour Millions PELOTON ADMISSIONS APP WITH THREE-DAY HORIZON Actual Predicted Ensemble (LR+RNN) Nov 30-Nov 4-Dec 8-Dec 12-Dec 16-Dec

40 Queries Per Hour Millions Millions PELOTON ADMISSIONS APP WITH THREE-DAY HORIZON Actual Predicted Ensemble (LR+RNN) Nov 30-Nov 4-Dec 8-Dec 12-Dec 16-Dec Hybrid (LR+RNN+KR) Nov 30-Nov 4-Dec 8-Dec 12-Dec 16-Dec

41 Design Considerations for Autonomous Operation

42 AUTONMOUS DBMS 19 DESIGN CONSIDERATIONS Configuration Knobs Internal Metrics Action Engineering

43 CONFIGURATION KNOBS UNTUNABLE KNOBS 20 Anything that requires a human value judgement should be marked as off-limits to autonomous components. File Paths Network Addresses Durability / Isolation Levels

44 CONFIGURATION KNOBS HOW TO CHANGE 21 The autonomous components need hints about how to change a knob Min/Max Ranges. Non-Uniform Deltas. Separate knobs to enable/disable a feature.

45 CONFIGURATION KNOBS HOW TO CHANGE 21 The autonomous components need hints about how to change a knob Min/Max Ranges. Non-Uniform Deltas. Separate knobs to enable/disable a feature.

46 CONFIGURATION KNOBS HARDWARE RESOURCES 22 Indicate which knobs are constrained by hardware resources. The sum of all buffers cannot exceed the total amount of available memory. The problem is that sometimes it makes sense to overprovision.

47 INTERNAL METRICS HARDWARE INFORMATION 23 Expose DBMS's hardware capabilities through a SQL interface. CPU, Memory, Disk, Network Configuration Recommender

48 INTERNAL METRICS SUB-COMPONENTS 24 If the DBMS has sub-components that are tunable, then it must expose separate metrics for those components.

49 INTERNAL METRICS SUB-COMPONENTS 24 If the DBMS has sub-components that are tunable, then it must expose separate metrics for those components.

50 INTERNAL METRICS SUB-COMPONENTS 24 If the DBMS has sub-components that are tunable, then it must expose separate metrics for those components.

51 INTERNAL METRICS SUB-COMPONENTS 24 If the DBMS has sub-components that are tunable, then it must expose separate metrics for those components.

52 ACTION ENGINEERING NO SHUTDOWN 25 No action should ever require the DBMS to restart in order for it to take affect. The commercial systems are much better than this than the open-source systems.

53 ACTION ENGINEERING NOTIFICATIONS 26 Provide a notification callback to indicate when an action starts and when it completes. Harder for changes that can be used before the action completes.

54 ACTION ENGINEERING RESOURCE USAGE 27 Support executing the same action with different resource usage levels.

55 ACTION ENGINEERING RESOURCE USAGE 27 Support executing the same action with different resource usage levels.

56 ACTION ENGINEERING RESOURCE USAGE 27 Support executing the same action with different resource usage levels.

57 ACTION ENGINEERING REJECTION HANDLING 28 Rejected actions should not cause DBMS to halt or unexpected application behavior. Network Threads Worker Threads Background Threads

58 ACTION ENGINEERING REJECTION HANDLING 28 Rejected actions should not cause DBMS to halt or unexpected application behavior. Network Threads Worker Threads Background Threads +2

59 ACTION ENGINEERING REJECTION HANDLING 28 Rejected actions should not cause DBMS to halt or unexpected application behavior. Network Threads Worker Threads Background Threads +2-1

60 ACTION ENGINEERING REJECTION HANDLING 28 Rejected actions should not cause DBMS to halt or unexpected application behavior. Network Threads Worker Threads Background Threads

61 ACTION ENGINEERING REJECTION HANDLING 28 Rejected actions should not cause DBMS to halt or unexpected application behavior. Network Threads Worker Threads Background Threads

62 ACTION ENGINEERING REPLICA EXPLORATION 29 Allow replica configurations to diverge from each other.

63 ACTION ENGINEERING REPLICA EXPLORATION 29 Allow replica configurations to diverge from each other. Master Replicas

64 ACTION ENGINEERING REPLICA EXPLORATION 29 Allow replica configurations to diverge from each other. Master Replicas

65 ACTION ENGINEERING REPLICA EXPLORATION 29 Allow replica configurations to diverge from each other. Master Replicas

66 What About Oracle's Self-Driving DBMS?

67 ORACLE SELF-DRIVING DBMS 31 September 2017 January 2017

68 ORACLE SELF-DRIVING DBMS 31 Automatic Indexing Automatic Recovery Automatic Scaling Automatic Query Tuning September 2017

69 ORACLE SELF-DRIVING DBMS 31 Automatic Indexing Automatic Recovery Automatic Scaling Automatic Query Tuning Problem #2 Reactionary Measures September 2017

70 ORACLE SELF-DRIVING DBMS 31 Automatic Indexing Automatic Recovery Automatic Scaling Automatic Query Tuning Problem #2 Reactionary Measures September 2017

71 CONCLUSION MAIN TAKEAWAYS 32 True autonomous DBMSs are achievable in the next decade. You should think about how each new feature can be controlled by a machine.

72