1 An Introduction to Real-Time Data Management 1
1. 1 A Brief History of Data Management 2
1. 2 Data Access: Pull vs. Push 4
1. 3 Query Semantics: Collections vs. Streams 5
1. 4 Chapter Outline 6
References 6
2 Database Management 9
2. 1 Triggers and Active Databases 9
2. 2 Change Data Capture, Cache Coherence, and Time-Series Data 10
2. 3 Materialized Views 12
2. 4 Change Notifications 13
2. 5 Summary and Discussion 14
References 14
3 Real-Time Databases 21
3. 1 What Is a Real-Time Database? 21
3. 2 What Is a Real-Time Query? 22
3. 3 System Landscape 24
3. 3. 1 Meteor 24
3. 3. 2 RethinkDB 28
3. 3. 3 Parse 29
3. 3. 4 Firebase 29
3. 3. 5 Baqend 31
3. 3. 6 Further Systems 35
3. 4 Summary and Discussion 36
References 38
4 Data Stream Management 43
4. 1 Queries Over Streams 43
4. 2 Notions of Time 46
4. 3 Windowing and Approximation 47
4. 4 Complex Event Processing 48
4. 5 Messaging Middleware 49
4. 6 Summary and Discussion 49
References 50
5 General-Purpose Stream Processing 57
5. 1 Architectural Patterns 57
5. 2 Batch vs. Stream Processing 60
5. 3 State-of-the-Art Stream Processing Frameworks 61
5. 3. 1 Storm 61
5. 3. 2 Samza 63
5. 3. 3 Spark Streaming 65
5. 3. 4 Flink 66
5. 3. 5 Further Systems 68
5. 4 Design Decisions and Trade-Offs 69
5. 5 Summary and Discussion 69
References 71
6 State of the Art and Future Directions 75
6. 1 The Big Picture 75
6. 2 Real-Time Databases: A New Paradigm for Data Management 76
6. 3 Closing Thoughts 77
Reference 77