OmniEdge Performance

OmniEdge v2.x is built entirely in Rust for memory safety, performance, and cross-platform compatibility. The core VPN transport is powered by OmniNervous, a WireGuard-based protocol implementation that provides industrial-grade network stability.

Industrial-Grade Stability

OmniEdge performance has been validated through 50-run longitudinal testing using Process Capability Analysis (Cpk):

What this means: Cpk > 2.0 indicates industrial-grade process capability. OmniEdge provides deterministic, jitter-controlled networking suitable for real-time robot control and latency-sensitive AI inference.

Test Setup

Tests were performed between 2 AWS EC2 m5.large instances in the same region.

m5.large Specifications

Intel Xeon Platinum 8000 CPU (64 bit)
2 vCPUs
8 GiB RAM
Up to 10 Gbps network
OS: Ubuntu 22.04 LTS

Test Tools

• iperf3 - Network throughput testing
• Custom latency measurement scripts with nanosecond precision
• 50-run statistical analysis for Cpk calculation

Why Cpk Matters

Process Capability Index (Cpk) is a statistical measure used in manufacturing to assess how well a process stays within specification limits. For networking:

OmniEdge achieves Cpk 2.92, meaning:

• Latency is highly predictable and stable
• Jitter is bounded within tight limits
• Suitable for deterministic applications like robot teleoperation

Use Cases Benefiting from High Stability

Robotics & Teleoperation

• Humanoid control loops: Deterministic latency for real-time control
• Robot swarms: Consistent mesh networking for multi-robot coordination
• ROS 2 DDS discovery: Reliable peer discovery across networks

AI & Machine Learning

• Federated learning: Consistent gradient exchange timing
• Distributed inference: Predictable model shard communication
• GPU cluster access: Stable connections to training infrastructure

Industrial IoT

• PLC remote access: Reliable connections to industrial controllers
• SCADA systems: Consistent monitoring data streams
• Predictive maintenance: Stable sensor data collection

NAT Traversal Performance (v2.1.0+)

OmniEdge includes advanced NAT traversal with minimal overhead:

Note: Relay adds ~10-50ms latency compared to direct P2P, but provides guaranteed connectivity for the most challenging NAT scenarios.

Protocol Comparison

Legacy Performance Data (v1.x)

For reference, here are the performance measurements from OmniEdge v1.x (Go/n2n):

Throughput on t2.micro (1 vCPU, 1 GiB)

Throughput on m5.large (2 vCPUs, 8 GiB)

Migration Note: OmniEdge v2.x significantly improves upon v1.x with the Rust rewrite and OmniNervous protocol, achieving better throughput and industrial-grade stability.

Conclusion

OmniEdge v2.x delivers:

• 484.7 Mbps throughput through encrypted tunnel
• 6-Sigma stability (Cpk 2.92) for deterministic networking
• ~0.3ms encryption overhead with WireGuard cryptography
• 99%+ NAT traversal success with automatic relay fallback

This makes OmniEdge ideal for robotics, AI, and industrial applications where network predictability is critical.

For detailed methodology and raw data, see the OmniNervous Capability Test Paper.

If you have more questions, feel free to discuss.