What Does It Enable?¶
Low Latency and More Compute for the Workload¶
By avoiding payload copies in the local data path, iceoryx2 minimizes the
CPU time and memory bandwidth spent on communication. At the same time,
message-delivery latency remains low and independent of payload size.
This results in sub-microsecond message-delivery latency, even for large payloads, while leaving more of the available compute resources for the algorithms and application logic that actually process the data.
Message-delivery latency with increasing payload size¶
For data-intensive systems, this allows larger workloads to run on the same hardware and helps scale applications from powerful development machines to more resource-constrained embedded targets.
Predictable Execution¶
Fast communication alone is not enough for real-time systems. Communication also needs to be predictable and avoid unexpected background activity that interferes with application execution.
iceoryx2 has no background threads, no system calls in the data-delivery
path, and no central broker. Message delivery therefore does not depend on
hidden middleware activity or additional scheduling decisions.
Together with the separation of data flow and control flow, this gives applications explicit control over when threads wake up, when data is consumed, and where execution takes place.
The result is stable low latency with minimal jitter and communication behavior that is easier to reason about under real-time constraints.
A Path Toward Mission-Critical Deployment¶
iceoryx2 is designed for demanding embedded environments where predictable
resource usage and explicit control over system behavior matter.
It supports no_std environments and avoids heap allocations during runtime.
Communication resources are bounded and configurable, while the decentralized
architecture avoids a central component that all communication depends on.
These properties make iceoryx2 a suitable communication foundation for
systems with real-time, robustness, and resource constraints.
They do not make a complete system safe or certified by themselves. But starting with a data plane designed for these constraints helps avoid replacing a fundamental part of the architecture when moving from prototype to production.
Further Reading¶
Learn how iceoryx2 keeps execution under application control.
Explore the message-delivery latency characteristics of iceoryx2.
Learn how iceoryx2 can be built for no_std environments.