Benchmark Node Throughput and Latency - The Messenger

                      TASK
                    

Implementation

How fast is your node? In production systems, you need to measure throughput (messages per second) and latency (time to process each message). Profiling reveals whether the bottleneck is in parsing, dispatch, or serialization.

Your task is to add benchmarking to your node:

Track the timestamp of each incoming message and each outgoing response
Compute per-message latency (time from receive to send)
Compute overall throughput (messages processed per second)
Report statistics via a bench_stats message type

Request:  {"type": "bench_stats", "msg_id": 1}
Response: {"type": "bench_stats_ok", "in_reply_to": 1, 
           "total_messages": 100, 
           "elapsed_ms": 523,
           "throughput_per_sec": 191.2,
           "avg_latency_us": 45,
           "p99_latency_us": 120}

Additionally implement a bench_echo type that is identical to echo but records timing:

Request:  {"type": "bench_echo", "msg_id": 1, "echo": "perf"}
Response: {"type": "bench_echo_ok", "in_reply_to": 1, "echo": "perf", "latency_us": 42}

Sample Test Cases

Init and echo still workTimeout: 5000ms

Input

{"src":"c0","dest":"n1","body":{"type":"init","msg_id":1,"node_id":"n1","node_ids":["n1"]}}
{"src":"c1","dest":"n1","body":{"type":"echo","msg_id":2,"echo":"bench"}}

Expected Output

{"src": "n1", "dest": "c0", "body": {"type": "init_ok", "in_reply_to": 1, "msg_id": 0}}
{"src": "n1", "dest": "c1", "body": {"type": "echo_ok", "echo": "bench", "in_reply_to": 2, "msg_id": 1}}

Bench echo includes latency_us fieldTimeout: 5000ms

Input

{"src":"c0","dest":"n1","body":{"type":"init","msg_id":1,"node_id":"n1","node_ids":["n1"]}}
{"src":"c1","dest":"n1","body":{"type":"bench_echo","msg_id":2,"echo":"perf"}}

Expected Output

{"src": "n1", "dest": "c0", "body": {"type": "init_ok", "in_reply_to": 1, "msg_id": 0}}

Hints

Hint 1▾

Track timestamps for each message received and each response sent

Hint 2▾

Throughput = total messages / elapsed time

Hint 3▾

Latency = time between receiving a message and sending the response

Hint 4▾

Use time.monotonic() for accurate elapsed time measurements

Hint 5▾

Store latency samples to compute percentiles (p50, p99)

Resources

Latency Numbers Every Programmer Should Know

Interactive visualization of latency at different system levels

                      OVERVIEW
                    

Theoretical Hub

Concept overview coming soon

Key Concepts

benchmarkingthroughputlatencyprofilingperformance

main.py

python

#!/usr/bin/env python3
import sys
import json
import time
class Node:
    def __init__(self):
        self.node_id = None
        self.node_ids = []
        self.next_msg_id = 0
        self.start_time = None
        self.message_count = 0
        self.latencies = []  # microseconds
    def send(self, dest, body):
        body["msg_id"] = self.next_msg_id
        self.next_msg_id += 1
        message = {"src": self.node_id, "dest": dest, "body": body}
        print(json.dumps(message), flush=True)
    def reply(self, request, body):
        body["in_reply_to"] = request["body"]["msg_id"]
        self.send(request["src"], body)
    def record_latency(self, start_us):
        # TODO: Compute and store latency in microseconds
        pass
    def get_percentile(self, percentile):
        # TODO: Compute the given percentile from self.latencies
        # e.g., p99 = 99th percentile
        pass
def main():
    node = Node()
    for line in sys.stdin:
        recv_time = time.monotonic()
        line = line.strip()
        if not line: