The Apache Pulsar team has announced the release of Apache Pulsar Client C++ version 4.0.0. Improvements include support for getting the encryption context on a message, and for getting the producer name of a message. The supported C++ standard has also been upgraded to version 17.
Apache Pulsar, originally developed by Yahoo, is an all-in-one messaging and streaming platform. Messages can be consumed and acknowledged individually or consumed as streams with less than 10ms of latency. Its layered architecture allows rapid scaling across hundreds of nodes, without data reshuffling.. Pulsar is multi-tenant, and provides a high-performance solution for server-to-server messaging.
Pulsar…
The Apache Pulsar team has announced the release of Apache Pulsar Client C++ version 4.0.0. Improvements include support for getting the encryption context on a message, and for getting the producer name of a message. The supported C++ standard has also been upgraded to version 17.
Apache Pulsar, originally developed by Yahoo, is an all-in-one messaging and streaming platform. Messages can be consumed and acknowledged individually or consumed as streams with less than 10ms of latency. Its layered architecture allows rapid scaling across hundreds of nodes, without data reshuffling.. Pulsar is multi-tenant, and provides a high-performance solution for server-to-server messaging.
Pulsar combines features of traditional message queues and streaming systems. One intersting aspect of Pulsar is that it has a layered architecture that separates the compute (brokers) from storage (BookKeeper). It supports microservices, real-time data pipelines, geo-replication, tiered storage, and built-in stream processing (Pulsar Functions).
Pulsar guarantees at-least-once delivery of messages, and offers automatic cursor management for subscribers. Its key features start with native support for multiple clusters in a Pulsar instance, with seamless geo-replication of messages across clusters. It combines very low publish and end-to-end latency, with seamless scalability to over a million topics. For programmers, there a simple client API with bindings for Java, Go, Python and C++. You can have multiple subscription types including exclusive, shared, and failover for topics; and message delivery is guaranteed with persistent message storage provided by Apache BookKeeper.
Topics (i.e., partitions) are divided among Pulsar brokers. A broker receives messages for a topic and appends them to the topic’s active virtual file (a.k.a ledger), hosted on the Bookkeeper cluster. Brokers read messages, mostly from the cache or alternatively from BookKeeper and dispatch them to the consumers. Pulsar and BookKeeper use Apache ZooKeeper to save metadata coordinated between nodes, such as a list of ledgers per topic, segments per ledger, and mapping of topic bundles to a broker.
A serverless lightweight computing framework, Pulsar Functions, offers the capability for stream-native data processing, and there’s a serverless connector framework, Pulsar IO, which is built on Pulsar Functions, that makes it easier to move data in and out of Apache Pulsar. Pulsar’s tiered storage offloads data from hot/warm storage to cold/long-term storage (such as S3 and GCS) when the data is aging out.
Pulsar Client C++ 4.0 is available now.
More Information
Related Articles
Cloudera And StreamNative Open Source NiFi Pulsar Connector
DataStax Astra DB gets Change Data Capture
Apache EventMesh Reaches Top Level Status
To be informed about new articles on I Programmer, sign up for our weekly newsletter,subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.
Comments
or email your comment to: comments@i-programmer.info