Platform results

Raspberry Pi 4

ARM64 Linux kernel implementation using the shared Rust core, platform-specific carrier adapter and Gigabit Ethernet.

ARM64Linux 6.18.38-stcp+Shared Rust coreTCP carrierUDP carrier

Implementation status

Custom ARM64 kernel boots
STCP kernel module loads
Directional crypto self-test
Remote carrier accept and handshake
End-to-end encrypted echo traffic
TCP benchmark dashboard published

Current published benchmark

Loading benchmark description…

Interim result: the full TCP and UDP carrier matrix is still running. The page reads JSON, so the final dataset can replace the interim file without editing HTML.

Combined throughput

Higher is better

Current observation

17.02%STCP/TCP client CPU in the 64 B test

STCP/TCP exceeded TLS throughput while using substantially less client CPU.

Detailed result

ModeCombined MiB/sOps/sRTT p50 msRTT p95 msRTT p99 msConnect msCPUErrors

Automatic benchmark analysis

Each matching test shape is compared automatically across TCP, TLS, STCP/TCP and STCP/UDP. Positive throughput and operations values mean STCP is faster; positive reduction values mean STCP uses less CPU or latency.

IRQ and kernel activity analysis

The benchmark captures Raspberry Pi server counters from /proc/interrupts, /proc/softirqs and /proc/stat before and after every test. Results are normalized per 1,000 successful operations.

Efficiency proxy, not watts: fewer network IRQ and softirq events per operation generally indicate less kernel processing work, but this graph is not a direct electrical power measurement.

Relative processing activity

The relative index uses measured network IRQ + NET_RX + NET_TX events per 1,000 operations. TLS is normalized to 100 whenever a matching TLS result exists.

Power and energy efficiency

The site distinguishes computational efficiency from actual electrical consumption. Real power comparisons appear only when measured watt or joule data is present.

Critical interpretation checklist

Raw speed

Combined throughput and operations per second by payload, client count and pipeline depth.

Latency quality

Median, p95 and p99 RTT. Tail latency matters for control systems and interactive device traffic.

Connection cost

Mean connection-establishment time, especially important for reconnecting IoT devices.

CPU efficiency

Throughput and operations normalized by client CPU percentage.

Reliability

Error counts, timeout rates and successful-operation ratios.

Scaling

Performance change from one client to high-concurrency tests and deeper pipelines.

Carrier comparison

STCP/TCP versus STCP/UDP under identical test shapes.

Energy efficiency

Measured watts, joules per MiB and joules per 1,000 operations when instrumentation data is available.

Integration coverage

MQTT

Suitable for secure telemetry and command channels from Raspberry Pi gateways.

MQTT section →

CoAP

Suitable for compact device APIs and constrained-resource access through STCP carriers.

CoAP section →

Raw socket API

Benchmark and application traffic use the standard socket-oriented STCP interface.