MicroStream: When a Java Application Becomes a DB

airhacks.fm podcast with adam bien

An airhacks.fm conversation with Markus Kett (@MarkusKett) about:

"What was your first computer?" - Markus was introduced in the episode #36, storing graph of Java objects with microstream, no annotation, not XML required, lazy subgraph loading, database support, coherence and cloud block storage (e.g. S3) are supported, microstream relies on key-value stores, using flat files, microstream relies on custom Java serialization, Java serialization challenges, microstream and security, microstream is not based on Java serialization, code execution during deserialization of Java objects is not avoidable, hackathlon with OracleLabs, Helidon and GraalVM, abstracting JVMs object ids, working with persistent Java objects directly, using getters for object traversal, working with Java object directly in memory, microstream can be orders of magnitudes faster than Java Persistence API, (JPA), accessing persistent object in microseconds, avoiding the JDBC IO- overhead, using Java's off-heap memory, persistent RAM and Intel's Optane, keeping Java object in RAM forever, thinking as Java developers, using Java collections as persistent objects, caffeine - the concurrent cache for Java, reasons for opensourcing microstream, long term support comes with commercial support, running microstream on GraalVM in native mode, polyglot persistence with GraalVM helidon is obsessed with performance, microstream on helidon on GraalVM, combining microstream and Kafka, kafka connector for microstream comes in the next release, microstream - redis integration, custom serialization formats, CDC and debezium, NoSQL database on top of microstream, object graph in Java is a multi-model database, the Java application becomes the database system, authorization on JPA object level, JPA security, the MicroStream, Helidon and GraalVM hackathlon, JAVAPRO magazine - the first free Java magazine, JCon is organized by JavaPRO,

Markus Kett on twitter: @MarkusKett