bloomberg/blazingmq

BlazingMQ is a distributed message queuing system that provides durable, fault-tolerant, and highly available message queues for asynchronous communication between application services.

The system solves the problem of enabling loosely-coupled communication between producers and consumers by storing messages until recipients retrieve and process them. This temporal and spatial isolation allows applications to operate independently. BlazingMQ implements various message routing strategies including work queues, priority routing, fan-out, and broadcast modes. The framework also includes features such as compression, strong consistency guarantees, and poison pill detection to handle edge cases in message processing.

BlazingMQ suits teams building distributed systems that require reliable message delivery and flexible routing patterns. The message broker is implemented in C++, with client libraries available in C++, Java, and Python, making it accessible across different technology stacks. Organizations with production workloads requiring high performance and fault tolerance will find the system's emphasis on efficiency and reliability relevant. The tool is particularly appropriate for teams already familiar with message queue concepts and seeking a system with strong consistency properties rather than eventual consistency trade-offs.

The project maintains active development with regular updates to the codebase. The tool has undergone extensive production validation over multiple years at a major financial institution, demonstrating stability in demanding environments. Documentation is comprehensive and includes quick-start guides with Docker-based local experimentation paths, as well as intermediate and advanced feature walkthroughs. The project accepts community contributions and maintains clear security vulnerability reporting procedures.