100% Free Forever
AI-Powered Learning
Industry Expert Content
Certificates & Badges
Learn At Your Own Pace
DevOps

Apache Pulsar

By the Apache Software Foundation

AdvancedPlatform4.5K learners

Apache Pulsar is a cloud-native, distributed messaging and streaming platform that unifies queuing and streaming in a single API, with built-in multi-tenancy, geo-replication, and storage that scales independently of compute.

Definition

Apache Pulsar is a cloud-native, distributed messaging and streaming platform that unifies queuing and streaming in a single API, with built-in multi-tenancy, geo-replication, and storage that scales independently of compute.

Overview

Pulsar was originally built at Yahoo to handle the company's large-scale messaging needs and was open-sourced before being donated to the Apache Software Foundation, where it graduated to a top-level project in 2018. Architecturally, Pulsar separates the serving layer from the storage layer: stateless brokers handle client connections and routing, while a dedicated storage layer persists data durably, allowing compute and storage to scale independently — a key difference from Apache Kafka, where brokers own both roles together. Pulsar also natively supports both traditional queue semantics and log-based streaming through one API, along with multi-tenancy for isolating workloads on shared infrastructure, geo-replication across data centers, and tiered storage that offloads older data to cheaper object storage. Pulsar Functions provide lightweight, built-in stream processing without standing up a separate compute cluster. It is commonly run on Kubernetes and packaged with Docker for cloud-native deployments, and is commercially supported by companies including StreamNative. Pulsar is generally positioned as a more operationally flexible, cloud-native alternative to Kafka for organizations that need multi-tenancy or geo-replication as first-class features rather than add-ons.

Key Features

  • Unified API for both message queuing and event streaming
  • Independent scaling of compute (brokers) and storage
  • Native multi-tenancy for isolating workloads
  • Built-in geo-replication across data centers
  • Tiered storage offloading older data to object storage
  • Pulsar Functions for lightweight built-in stream processing
  • Schema registry for enforcing message formats

Use Cases

Financial services and payment messaging requiring strict isolation
Multi-region event streaming with geo-replication
Microservices communication in cloud-native architectures
IoT and telemetry ingestion at scale
Real-time analytics pipelines alongside batch processing

Frequently Asked Questions