Apache Pulsar is an open-source cloud-native streaming and messaging platform. Its multi-layer architecture enables organizations to build durable, scalable, reliable applications in elastic cloud environments. Thousands of companies rely on Pulsar for high-performance data pipelines, event-driven microservices, real-time analytics, and other real-time workloads. Pulsar was originally developed at Yahoo, and committed to open source in 2016.
Apache Kafka is an open-source monolithic streaming platform. Kafka is most commonly used by companies for streaming use cases such as high-performance data pipelines and streaming analytics. Kafka was first developed at LinkedIn and open sourced in 2011.
Pulsar provides a multi-layer architecture that decouples storage and compute. Pulsar’s design allows organizations to elastically scale storage independently from compute and achieve different levels of resource isolation. In contrast, Kafka has a monolithic architecture that tightly couples compute and storage, where resources must be scaled together.
Pulsar has the flexibility to serve a wide range of messaging and event streaming use cases in one platform. Users can deploy Pulsar as a centralized messaging platform as well as for batch and event streaming. In comparison, Kafka primarily targets event streaming use cases, with only limited messaging support.
In our 2022 test comparing the performance of Pulsar and Kafka, we found:
Pulsar vs. Kafka features comparison