Acknowledged replication of Queue across the network

Benchmarks have a natural lifespan that can be improved with more modern hardware. This benchmark was performed 18 March 2020.  Please don’t hesitate to contact us if you have questions about how Chronicle Queue Enterprise will perform today.

A major feature of Chronicle Queue Enterprise is support for TCP and UDP replication across multiple servers to ensure high availability of client infrastructure, whilst there may be inherent latencies within the network, we want to ensure that Queue replication adds minimal overheads. Below is a chart which shows how long it takes to write a 40 byte message then write replicate over TCP have a second acknowledged receipt and finally have the thread read the acknowledged message. The results show that typical latencies varied between 14 and 40 microseconds and the 99th percentile varied between 17 and 56 microseconds dependent on throughput being tested

Latency distribution

Andrew Twigg

Subscribe to Our Newsletter

Featured Products

Data Processing

Chronicle Queue Enterprise

Persisted messaging framework that handles massive throughput. Delivers superior control of outliers, and replicates easily across hosts.