RTI Data Distribution Service

Latency

These graphs show the average one-way latency in microseconds for publish/subscribe messaging. Latency was measured by having the consumer (DDS DataReader) echo messages back to the producer (DDS DataWriter). This allowed roundtrip latency to be measured on the sending machine, avoiding time synchronization issues. The roundtrip latency was divided in half to get the one-way latency that is shown.

These results show that RTI's latency remains consistently low at small message sizes, starting at 71 microseconds. At larger messages sizes, which are network-limited, latency increases proportionally to message size.

Jitter

This graph is based on the same data as the previous latency graphs. It shows the same average latency (in red) but adds jitter—the variation in latency from message to message. A system is more deterministic if it exhibits lower jitter.

The two blue series show the minimum measured latency and the 99% latency (the latency below which 99% of the samples fell). Except at the largest message sizes, the variation between minimum and 99% latency is consistently under 20 microseconds. This shows that RTI Data Distribution Service exhibits very low jitter and very high determinism, making it suitable for time- and mission-critical applications.

Throughput - Bandwidth

This graph shows sustainable one-to-one (point-to-point) throughput in terms of network bandwidth (megabits per second). Throughput was measured in two scenarios:

• between single producing and consuming threads (shown in red)
• between four pairs of producing and consuming threads, with the producers running on one quad core processor and the consumers on another (shown in green)

Accounting for Ethernet, IP and UDP overhead, the maximum bandwidth available for message data (and metadata) is slightly over 950 megabits. With four threads, RTI Data Distribution Service is able to fully utilize all of this available bandwidth when sending messages larger than 128-bytes. With a single thread, RTI is able to take advantage of the available bandwidth above 512 bytes. Beyond these points, throughput is limited by the network and not by the CPU or RTI Data Distribution Service.

Because RTI Data Distribution Service uses true peer-to-peer messaging—with no centralized or per-node Enterprise Service Bus (ESB), message broker, server or daemon processes—it does not impose any inherent limit on aggregate messaging capacity. It is limited only by the network infrastructure. For an illustration of this, see the Topic and Scalability benchmark.

Throughput - Message Rate

These graphs show throughput in terms of the message rate (measured in messages per second). At small message sizes, a single thread can send or receive nearly 750,000 messages per second. Four parallel threads running on the same quad core processor and sharing the same NIC can send or receive almost 1,750,000 messages per second. Beginning around 512-byte messages, throughput is the same regardless of the number of producer-consumer pairs since the network is saturated.