Practical Real-time Data Processing and Analytics

Aggregator

The following are the characteristics of aggregator:

• It emits any number of tuples with any number of fields.
• It executes the init method before processing the batch.
• It executes the aggregate method for each input tuple in the batch partition. This method can update the state and optionally emit tuples.
• It executes the complete method when all tuples for the batch partition have been processed by the aggregate method.
• Interface available in package:

public interface Aggregator<T> extends Operation {  T init(Object batchId, TridentCollector collector);  void aggregate(T state, TridentTuple tuple, TridentCollector collector);  void complete(T state, TridentCollector collector);}

Example:

public class CountAgg extends BaseAggregator<CountState> ...