Code samples for Hazelcast Jet using the DAG API.
This sample shows, how to enrich items with additional information by matching them by key
Two samples that demonstrate sliding window aggregation in a single-stage and in a two-stage setup.
Shows how to implement a custom distributed sink that stores the data in files.
Shows how to implement a custom distributed source, including custom
partitioning at the source using the ProcessorMetaSupplier
API.
The sample implements a distributed generator of integers which is
used as the source for a filtering vertex that selects the prime
numbers from it.
Demonstrates the power of the Core API by building a hand-optimized DAG that cannot be reproduced with the higher-level APIs. The sample builds an inverted index on a corpus of about a 100 MB of book material and then presents you with a GUI dialog where you can enter your search terms. The GUI poignantly demonstrates the speed of the search by instantly responding to every keystroke and displaying a result list.
The classical Word Count task implemented in the Core API.