Why does the total query execution time grow too fast as the number of queries increases?

Question

Python

I use the Wandbox API in my project , sending requests asynchronously.

The code:

import time
import aiohttp
import asyncio

URL = 'https://wandbox.org/api/compile.json'
HEADERS = {'Content-Type': 'application/json'}

code = '''
class Compiler{
    public static void main(String[] args){
        System.out.println("This works!");
    }
}
'''

data = {
    'code': code,
    'compiler': 'openjdk-head',
    'stdin': ''
}


async def post(session):
    async with session.post(url=URL, headers=HEADERS, json=data) as wandbox_response:
        result = await wandbox_response.json()
        return result


async def gather_tasks(N):
    async with aiohttp.ClientSession() as session:
        tasks = []

        for i in range(N):
            tasks.append(asyncio.ensure_future(post(session=session)))

        result = await asyncio.gather(*tasks)

    return result


for number_of_tasks in range(1, 10):
    start_time = time.time()

    loop = asyncio.get_event_loop()  # Creating async loop
    results = loop.run_until_complete(gather_tasks(number_of_tasks))

    print(number_of_tasks, time.time() - start_time)

If you run it, then the console will display:


If you change the compiler, for example, to python and send the corresponding code:

code = '''
print('This works!')
'''

data = {
    'code': code,
    'compiler': 'cpython-head',
    'stdin': ''
}

Then the console will display:


Why is this happening and how to fix it?