Multi-Protocol Event Broker

Async Event Broker & Queue
With Zero Overhead

Pranor Pulse unifies Kafka stream processing, RabbitMQ queues, STOMP WebSockets, and IoT MQTT protocols into a single Go binary with real-time Dead Letter Queue (DLQ) replay.

Quick Start View on GitHub

Standalone Install: docker run -p 8083:8083 vyuvaraj/pranor-pulse:latest

Capabilities

One Broker for All Messaging Workloads

Stop running separate JVMs for Kafka, Erlang daemons for RabbitMQ, and Mosquitto brokers for IoT.

🔌

Multi-Protocol Protocol Engine

Native wire-protocol decoders for Kafka wire format, STOMP over WebSockets, and MQTT 3.1.1/5.0. Use your existing client libraries without modification.

🔄

Dead Letter Queue (DLQ) Replay

Failed messages automatically route to topic DLQs with configurable exponential backoff. Trigger one-click message replays via API or VS Code extension.

📊

Native Prometheus Metrics

Built-in OTLP metrics exporter publishing message throughput, lag, consumer offsets, and connection state directly to Grafana dashboards.

Comparison Matrix

Pulse vs Kafka & RabbitMQ

Compare resource requirements and protocol support.

Feature	Kafka / RabbitMQ	Pranor Pulse
Runtime Dependencies	✕ JVM / Erlang VM	✓ Zero (Native Go Binary)
Multi-Protocol Support	✕ Requires separate clusters	✓ Kafka + STOMP + MQTT built-in
1-Click DLQ Message Replay	✕ Complex script setup	✓ Built-in API & Webview
Baseline RAM Usage	500MB - 2GB+	✓ ~18MB
Embedded Mode Support	✕	✓ Can be embedded in Go apps
Open Source (AGPL-3.0)	✓	✓

Code Examples

Use from Any Language or Framework

Pranor Pulse supports Python (Kafka/STOMP), TypeScript/Node.js, Go, cURL/WebSockets, or directly in Pranor applications.

from kafka import KafkaProducer, KafkaConsumer
import json

# 1. Connect standard Python Kafka client to Pranor Pulse
producer = KafkaProducer(bootstrap_servers=['localhost:8083'], value_serializer=lambda v: json.dumps(v).encode('utf-8'))
producer.send('orders.created', {'order_id': 'ord_9901', 'amount': 149.50})

# 2. Consume events
consumer = KafkaConsumer('orders.created', bootstrap_servers=['localhost:8083'])
for msg in consumer:
    print("Received order event:", msg.value)
  

import { Kafka } from "kafkajs";

// Standard KafkaJS client pointed at Pranor Pulse broker
const kafka = new Kafka({ clientId: "node-app", brokers: ["localhost:8083"] });
const producer = kafka.producer();
await producer.connect();
await producer.send({
  topic: "orders.created",
  messages: [{ key: "ord_9901", value: JSON.stringify({ amount: 149.50 }) }]
});
  

package main

import "github.com/segmentio/kafka-go"

func main() {
    writer := &kafka.Writer{
        Addr:     kafka.TCP("localhost:8083"),
        Topic:    "orders.created",
    }
    writer.WriteMessages(context.Background(), kafka.Message{
        Key:   []byte("ord_9901"),
        Value: []byte(`{"amount": 149.50}`),
    })
}
  

# Publish event via Pulse HTTP Bridge
curl -X POST http://localhost:8083/api/v1/topics/orders.created/publish \
  -H "Content-Type: application/json" \
  -d '{"key": "ord_9901", "payload": {"amount": 149.50}}'

# WebSocket event stream subscription
wscat -c ws://localhost:8083/events?topic=orders.created
  

import "std/pulse"

fn main() {
    // Native Pranor syntax
    pulse.publish("orders.created", "ord_9901", { "amount": 149.50 })
}
  

Async Event Broker & Queue With Zero Overhead