Stream ingestion and pub/sub systems - CS 591 K1: Data Stream Processing and Analytics Spring 2020Asynchronous communication, i.e. producer only needs to receive ack from broker 9 Communication patterns (I) Load balancing or shared subscription • A logical producer/consumer can be implemented by The broker can choose to send messages to consumers in a round-robin fashion 10 Communication patterns (II) Fan-out Several logical consumers (possibly implemented by several parallel physical processes) and basic comparison operators. • Constraints can be logically combined to form complex event patterns. • company == ‘Uber’ and price < 100 • Predecessors of Complex Event Processing (CEP) systems0 码力 | 33 页 | 700.14 KB | 1 年前3
Streaming languages and operator semantics - CS 591 K1: Data Stream Processing and Analytics Spring 2020conditions and actions to be taken when conditions are met. • Conditions are commonly described as patterns that can match input stream events on type, content, timing constraints. • Actions define how University 2020 Summary Today you learned: • there are various types of languages for data streams • patterns, transformations, declarative • traditional blocking operators don’t work on streams • non-blocking0 码力 | 53 页 | 532.37 KB | 1 年前3
Course introduction - CS 591 K1: Data Stream Processing and Analytics Spring 2020anomaly detection framework for datacenters. Your framework will: • Detect “suspicious” event patterns • Raise alerts for abnormal system metrics • Detect invariant violations • Identify outlier tasks0 码力 | 34 页 | 2.53 MB | 1 年前3
Streaming optimizations - CS 591 K1: Data Stream Processing and Analytics Spring 2020maximum every 100 events? • clicks per user session? • faster than the batch size? • alerts when patterns occur? 50 t t+1 t+3 t+4 t+5 t+6 t+7 t+2 How would you compute… ??? Vasiliki Kalavri | Boston0 码力 | 54 页 | 2.83 MB | 1 年前3
共 4 条
- 1













