Chronicle Queue Enterprise
Chronicle Queue Enterprise is a high-performance, low-latency messaging framework trusted by organisations requiring reliable and efficient data processing. The Enterprise edition enhances interoperability by eliminating multi-language compatibility barriers, enabling seamless data exchange across Java, C++, Rust and Python environments.
Talk to Our TeamKey Features
Ultra-Low Latency & High Throughput
Process millions of messages per second with consistently low latency, down to nanoseconds, depending on configuration and setup. Tail percentiles stay tightly controlled even as volumes spike.
Built-in HA/DR
Queue Enterprise Replication copies every message from a source queue to one or more sinks over TCP/IP. Another host holds the same messages in the same sequence, which is what lets a service resume without gaps.
Cross-Language Compatibility
Java, C++, Rust and Python libraries share a binary-compatible format and reads the same queues. Teams working in different languages exchange data directly, with no bridging layer and no rewrite.
Persisted Messaging
Every message is appended to disk in sequence and never overwritten. Recovery preserves the exact order written, creating a defensible audit trail that helps meet regulatory requirements.
Eliminate GC and Jitter
Chronicle Queue Enterprise manages memory off-heap, so no garbage collection pause lands in your critical path. Async Mode adds a buffer between your application and the operating system, flattening the tail latencies that appear under sustained load.
Expert Support & Tools
Direct access to Chronicle's team of experienced low-latency developers, with tools such as Chronicle Tune to help identify performance bottlenecks, diagnose latency issues & optimise and resolve system performance issues.
Technical Information
Explore the advanced engineering principles that shape this product’s core, ensuring exceptional performance, operational resilience, and seamless scalability
Chronicle Queue Enterprise
Chronicle Queue Enterprise is a low-latency message queue that persists messages in high-performance, mission-critical applications, engineered for exceptional performance, operational resilience and seamless scalability.
What is Chronicle Queue Enterprise?
A "record everything" message queue that operates at microsecond real-time latency. It behaves like a broker-less, durable, persisted JVM topic: in optimised scenarios, messages pass between JVMs on the same machine in under a microsecond, and between machines via replication in under 10 microseconds.
Read MoreAsync Mode
A multi-writer, multi-reader buffered queue that sits between your application and the queue on disk, absorbing the pauses the operating system introduces under load. Use it to flatten tail latencies and keep the critical path deterministic when the system is under sustained pressure.
Read MoreQueue Replication
Copies every message from a source queue to one or more sinks over TCP/IP on-premises, across data centers, in the cloud or in order, with acknowledgement available to the writer or the reader. Each sink holds the same messages in the same sequence, the condition that makes replay consistent across every host in the cluster.
Read MoreAdvanced
The engineering behind the latency figures. An append only structure suits sequential disk access, and memory mapped files give fast, thread safe access beyond JVM memory limits, moving the size ceiling from heap to available disk. Replaying 100 TB becomes a storage question.
Read More
Use Cases
How financial institutions run Chronicle Queue Enterprise in production against real integration, latency and scalability problems.
Why Chronicle Software?
Four languages, one binary format
Java, C++, Rust and Python libraries are all fully binary compatible. A Python research process and a C++ execution process read the same queue with no translation layer between them.
80%+ of the world's top 100 banks
Chronicle Software has been in production since 2013, with more than 200,000 annual global users across banks, exchanges, crypto firms, trading houses and hedge funds.
$10bn+ notional trading per day
That is the volume running through Chronicle's technology platform.
Multilanguage Support
Chronicle Queue Enterprise is available as a Java, Rust, C++, or Python library, with full binary compatibility across all versions.
FAQs
Get answers to frequently asked questions about Chronicle Queue Enterprise.
How Does Chronicle Queue Enterprise Work?
Chronicle Queue Enterprise appends every message to a memory-mapped file on disk, in sequence, and never resizes an existing entry. Because the file is memory-mapped, readers and writers in separate processes share it directly rather than passing objects through a broker, removing the network hop and the broker state machine from the path.
Data is held off-heap, so the garbage collector never sees it, and the append-only structure gives total event ordering, a reader replays the queue in the exact order it was written. Capacity is bounded by available disk rather than main memory, so a slow consumer never throttles the producer.
Does Chronicle Queue Enterprise Help With Regulatory Compliance?
Yes. Because every message is appended to disk in sequence and never overwritten, firms can always reconstruct exactly what happened and when. That's the deterministic, tamper-evident record regulators like MiFID II, CFTC and the SEC expect. Encryption adds a further layer of protection on top.
What Is Chronicle Queue Enterprise?
Chronicle Queue Enterprise is a high-performance, persisted messaging solution designed for applications where low and predictable latency, reliability and data integrity are critical.
It combines messaging and durable storage in a single solution, enabling applications to exchange, retain and replay large volumes of data while preserving event order. Supporting asynchronous RMI and publish/subscribe interfaces, it is designed for demanding use cases such as trading systems, market data, event-driven applications and other latency-sensitive workloads.
What Other Chronicle Solutions Work Alongside Chronicle Queue Enterprise?
Chronicle Queue Enterprise integrates with the broader Chronicle stack:
Chronicle FIX is a Java FIX engine that holds New Order Single round-trip latency under 4 microseconds at the 99th percentile. Use it where order flow arrives over FIX and has to be parsed, journaled and routed without adding milliseconds.
Chronicle Queue Backed Map is a replicated key-value store implementing the standard Java ConcurrentMap API, with Chronicle Queue as its unit of replication. Every map write is committed first as an immutable queue excerpt, giving a complete, ordered, replayable history of state transitions — and full map state is captured in a single queue event, so a restart does not trigger unbounded replay.
Chronicle Ring is a multi-writer, multi-reader ring buffer that reduces tail latencies and eliminates data copying. Chronicle publishes latencies as low as ~150 ns at the 50th percentile, holding under 800 ns at the 99.999th on standard enterprise hardware. Its RingZero variant is optimised for single-writer, single-reader traffic.
Chronicle Queue Enterprise also works alongside Chronicle’s wider suite of low-latency solutions. Chronicle’s consultants have tools available to improve the performance in your specific environment and are able to provide guidance on how best to use our solutions to improve performance.