I decided not to complicate my tests with additional filters, because when whatsapp philippines number looking at the usage logs, I noticed that most user queries did not have any filters. With this list, I created a short Python function that executes these queries and saves the duration of each of them in a dictionary results : Python Copy the code import random import subprocess from timeit import timeit requests = [ # ... ] results = {url: [] for url in requests} def test(server, apikey, start, end): requests_copy = requests[:] random.shuffle(requests_copy) for url in requests_copy: t = timeit(lambda: subprocess.
check_call( ['curl', '-f', f'{server}{url}&start={start}&end={end}', '-H', f'Authorization: Bearer {apikey}'], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL, ), number=1) results[url].append(t) The function test()makes a copy of the list requests, reorganizes it randomly, and then executes the request with curlas a subprocess. It uses the timeit()Python standard library function to measure the time it takes for each request to return a response, and then adds this measurement to the dictionary results, under the corresponding URL. The reason I randomize the list is that I intend to run multiple instances of this function in parallel to simulate concurrent clients. It is convenient for the function to iterate over the queries in random order, as this ensures that the database has a variety of queries at any given time, rather than receiving multiple instances of the same query.