Code samples for Hazelcast Jet using the Core API.
-
Shows how to implement a custom distributed sink that stores the data in files.
-
Migrating from Hazelcat Map-Reduce
Contains parallel samples with Hazelcast's Map-Reduce API and Hazelcast Jet's Core API that help you migrate your Map-Reduce code to Hazelcast Jet.
-
Shows how to implement a custom distributed source, including custom partitioning at the source using the
ProcessorMetaSupplierAPI. 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. -
Inverted Index with TF-IDF Scoring
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.
-
Shows how to use the File Watcher streaming source. It continuously monitors HTTP access log files for new content and applies a sliding window aggregation that tracks the frequency of visits to each page.
-
Illustrates the effects of different processing guarantees that a Jet job can be configured with. Uses a Hazelcast
IMap's Event Journal to perform rolling average calculations. -
Demonstrates the session window vertex to track the behavior of the users of an online shop application.
-
Two samples that demonstrate sliding window aggregation in a single-stage and in a two-stage setup.
-
Demonstrates cascaded sliding windows where the second one's source is the output of the first one. The first one calculates the frequency of trading each stock and the second one finds the hottest-trading stocks.