Speed Up Web Scraping with Concurrency in Python

May 23, 2024 · 10 min read

Scraping websites for data is a typical activity for developers. Whether it's a side project or you're building a startup, there are many reasons to scrape the web.

For example, if you want to start a price comparison platform, you'll need to retrieve prices from various e-commerce sites. You may want to build an AI to identify products and look up their cost on Amazon. The possibilities are endless.

But have you ever noticed how slow it is to get all the pages? Would you scrape the products one after the other? There must be a better solution, right? Right?!

Crawling websites can be time-consuming because you have to deal with rate-limiting and wait for server responses. Today, you'll learn how to speed up your web scraping projects by using concurrency in Python.

Let's start!


Here's what you need to follow this tutorial:

You must have Python 3 on your device. Note that some systems have it pre-installed. After that, add all the necessary libraries by running pip install.

pip install requests beautifulsoup4 aiohttp numpy

If you know the basics behind concurrency, skip the theory part and jump directly into the action.


Concurrency is a term that deals with the ability to run multiple computing tasks simultaneously.

When you make sequential requests to websites, you send out one at a time, wait for it to return, and then send out the following one.

However, with concurrency, you can send out many of these at once and work on all of them as they return. The speed boost is incredible! Compared to sequential requests, concurrent ones will be much faster regardless of whether they are running in parallel (multiple CPUs) or not. But more on this later. Let's see what the benefits of concurrency are:

To understand this, we need to examine the difference between processing tasks sequentially and concurrently. Let's do so with an example:

Say we have five tasks that take 10 seconds each to complete.

If we process them sequentially, we'll need 50 seconds to complete all five. However, it'll take only 10 seconds for all if we choose to proceed concurrently.

Click to open the image in full screen

In addition to increasing speed, concurrency allows us to do more work in less time by distributing our scraping workload among several processes.

There are several ways to parallelize requests, e.g., multiprocessing and asyncio. From a scraping perspective, we can use these libraries to parallelize requests to different websites or other pages on the same one.

In this article, we'll focus on asyncio, a Python module providing infrastructure for writing single-threaded concurrent code using coroutines.

Since concurrency implies more convoluted systems and code, consider if the pros outweigh the cons for your use case.

Benefits of Concurrency

  • More work done in less time.
  • Idle network time invested in other requests.

Dangers of Concurrency

  • Harder to develop and debug.
  • Race conditions.
  • The need to check and use thread-safe functions.
  • Block probabilities grow if not handled carefully.
  • Concurrency comes with a system overhead; you should set a reasonable level.
  • Involuntary DDoS if too many requests against a small site.
Frustrated that your web scrapers are blocked once and again?
ZenRows API handles rotating proxies and headless browsers for you.
Try for FREE

Why Asyncio?

To decide what technology to use, we must first understand the difference between asyncio and multiprocessing:

  • Asyncio is "a library used to write concurrent code using the async/await syntax." It runs on a single processor.
  • Multiprocessing is "a package that supports spawning processes using an API [...] allowing the programmer to fully leverage multiple processors on a given machine." Each process will start its own Python interpreter in a different CPU.

Also, how do I/O-bound and CPU-bound differ from each other?

  • I/O-bound means the program will run slower due to input/output operations. In our case, mostly network requests.
  • CPU-bound means the program will run slower due to central processor use, e.g., math calculations.

Why does this affect our choice?

A big part of the concurrency cost is creating and maintaining threads/processes. For CPU-bound problems, having many of those in different CPUs will pay off. But that might not be the case for I/O-bound scenarios.

Since scraping is mostly I/O-bound, we'll pick asyncio. But in case of doubt (or just for fun), you can replicate the idea using multiprocessing and compare the results.

Click to open the image in full screen

Sequential Version

We'll start by extracting data from ScrapingCourse.com, a demo e-commerce site that is the perfect playground to test our skills.

Let's first explore the sequential version of the scraper. Several snippets are part of all cases, so those will remain the same.

By visiting the website, we see that it has 48 pages. Our first constants will be the base URL and a range for the pages.

base_url = "https://www.scrapingcourse.com/ecommerce/page" 
pages = range(1, 13) # max page (48) + 1

Let's extract some data!

We'll use requests.get to get the HTML and then BeautifulSoup to parse it. We'll loop over each product and get some basic info from them. All the selectors come from a manual review of the content (using DevTools), but we won't go into detail here for brevity.

import requests 
from bs4 import BeautifulSoup 
base_url = "https://www.scrapingcourse.com/ecommerce/page" 
pages = range(1, 13) # max page (12) + 1

def extract_details(page): 
	# concatenate page number to base URL 
	response = requests.get(f"{base_url}/{page}/") 
	soup = BeautifulSoup(response.text, "html.parser") 
	product_list = [] 
	for product in soup.select(".product"): # loop each product 
			"name": product.find("h2").text.strip(), 
			"price": product.find(class_="price").text.strip(), 
			"url": product.find(class_="woocommerce-loop-product__link").get("href"), 
	return product_list

The extract_details function will take a page number and concatenate that to create a URL with the base. After getting the content and creating an array of products, return them. That means the returned values will be a list of dictionaries. This is an essential detail for later.

We need to run the function for each page, get all the results, and store them.

# ...
import csv 
# modified to avoid running all the pages unintentionally 
pages = range(1, 3) 
def store_results(product_list): 
	with open("products.csv", "w") as product_file: 
		# get dictionary keys for the CSV header 
		fieldnames = product_list[0].keys() 
		file_writer = csv.DictWriter(product_file, fieldnames=fieldnames) 
product_list = [ 
	for page in pages 

Running the code above will get two product pages, extract products (32 total), and store them in a CSV file called product.csv. The store_results function doesn't affect the scraping in sequential or concurrent mode. You can skip it.

Since the results are lists, we must flatten them to allow writerows to do its job. That's why we named the variable product_lists (even if it's a bit weird), as a reminder that it's not flat.

Example of the output CSV file:

name price url
Abominable Hoodie $69.00 https://www.scrapingcourse.com/ecommerce/product/abominable-hoodie/
Adrienne Trek Jacket $57.00 https://www.scrapingcourse.com/ecommerce/product/adrienne-trek-jacket/
Aeon Capri $48.00 https://www.scrapingcourse.com/ecommerce/product/aeon-capri/
Aero Daily Fitness Tee $24.00 https://www.scrapingcourse.com/ecommerce/product/aero-daily-fitness-tee/
Aether Gym Pant $74.00 https://www.scrapingcourse.com/ecommerce/product/aether-gym-pant/

If you were to run the script for every page (12 total), it would generate a CSV with 188 products in about 30 seconds.

time python script.py 
real 0m31,806s 
user 0m1,936s 
sys 0m0,073s

Introducing Asyncio

We know we can do better. If we perform all the requests at the same time, it should take much less, right? Maybe as long as the slowest request?

Concurrency should indeed run faster, but it also involves some overhead. So it isn't a linear mathematical improvement. But improve, we will.

Let's see how asyncio works!

It allows us to run several tasks on the same thread in an event loop (like JavaScript does). It'll run a function and switch the context to a different one when possible. In our case, HTTP requests allow that switch.

We should see an example that sleeps for a second. And a second is all it takes for the script to run. Notice that we cannot call main directly. We need to let asyncio know that it's an async function that needs executing.

import asyncio 
async def main(): 
	print("Hello ...") 
	await asyncio.sleep(1) 
	print("... World!") 
time python script.py 
Hello ... 
... World! 
real 0m1,054s 
user 0m0,045s 
sys 0m0,008s

Simple Code in Parallel

Next, we will expand the example case to run a hundred functions. Each of them will sleep for a second and print a text. It'd take around 100 seconds if we were to run them sequentially. With asyncio, it will take just one!

That's the power behind concurrency. As said earlier, for pure I/O-bound tasks, it will perform much faster (sleeping is not, but it counts for the example).

We need to create a helper function that'll sleep for a second and print a message. Then, we edit main to call it a hundred times and store each call in a task list. The last and crucial part is to execute and wait for all the tasks to finish. That's what asyncio.gather does.

import asyncio 
async def demo_function(i): 
	await asyncio.sleep(1) 
	print(f"Hello {i}") 
async def main(): 
	tasks = [ 
		for i in range(0, 100) 
	await asyncio.gather(*tasks) 

As expected, a hundred messages and one second to execute. Perfect!

time python script.py 
Hello 0 
Hello 99 
real 0m1,065s 
user 0m0,063s 
sys 0m0,000s

Scraping with Asyncio

We need to apply that knowledge to scraping. The approach to follow will be to request concurrently and return product lists. Once all requests finish, store them. It might be better to save information after each request or in batches to avoid data losses for real-world cases.

Our first attempt won't have a concurrency limit, so be careful when using it. In the case of running it with thousands of URLs... well, it'd perform all of those almost at the same time. Which could cause a tremendous load on the server and probably fry your computer.

requests does not support async out-of-the-box, so we will use aiohttp to avoid complications. requests can do the job, and there is no substantial performance difference. But the code is more readable using aiohttp.

import asyncio 
import aiohttp 
from bs4 import BeautifulSoup 
async def extract_details(page, session): 
	# similar to requests.get but with a different syntax 
	async with session.get(f"{base_url}/{page}/") as response: 
		# notice that we must await the .text() function 
		soup = BeautifulSoup(await response.text(), "html.parser") 
		# [...] same as before 
		return product_list 
async def main(): 
	# create an aiohttp session and pass it to each function execution 
	async with aiohttp.ClientSession() as session: 
		tasks = [ 
			extract_details(page, session) 
			for page in pages 
		list_of_lists = await asyncio.gather(*tasks) 

The CSV file should have every product (755), just as before. Since we perform all the page calls simultaneously, the results won't arrive in order. If we were to add the results to the file inside extract_details, they might be unordered. Since we wait for all tasks to finish and then process them, this won't be a problem.

time python script.py 
real 0m11,442s 
user 0m1,332s 
sys 0m0,060s

We did it! 3x faster is nice, but... shouldn't it be 40x? It's not that simple. Many things can affect the performance, e.g., network, CPU, and RAM.

We've noticed that response time slows down in the demo when we perform several calls. It might be by design. Some servers/providers can limit the number of concurrent requests to avoid too much traffic from the same IP. This isn't a block but more of a queue.

To see real speed-up, you can test against a delay page. It's another testing page that will wait for two seconds and then return a response.

base_url = "https://httpbin.org/delay/2" 
async def extract_details(page, session): 
	async with session.get(base_url) as response: 

Removed all the extracting and storing logic, just calling the delay URL 48 times. And it runs in under three seconds.

time python script.py 
real 0m2,865s 
user 0m0,245s 
sys 0m0,031s

Limiting Concurrency With Semaphore

As said earlier, we should limit the number of concurrent requests, especially against a single domain.

asyncio comes with Semaphore, an object that'll acquire and release a lock. Its inner functionality will block some of the calls until the lock is acquired, thus enabling maximum concurrency.

We need to create the semaphore with the maximum we want. And then wait until the extracting function is available using async with sem.

max_concurrency = 3 
sem = asyncio.Semaphore(max_concurrency) 
async def extract_details(page, session): 
	async with sem: # semaphore limits num of simultaneous downloads 
		async with session.get(f"{base_url}/{page}/") as response: 
			# ... 
async def main(): 
		# ... 
loop = asyncio.get_event_loop() 

It gets the job done, and it is relatively easy to implement! Here's the output with a max concurrency of three:

time python script.py 
real 0m13,062s 
user 0m1,455s 
sys 0m0,047s

This shows that the version with unlimited concurrency isn't operating at its full speed 🤦. If we increment the limit to 10, the total time is similar to the unbound script.

Limiting Concurrency With TCPConnector

aiohttp offers an alternative solution with further configuration. We can create the client session passing in a custom TCPConnector.

We can build it by using two parameters that suit our needs:

  • limit: total number of simultaneous connections.
  • limit_per_host: limit simultaneous connections to the same endpoint (same host, port, and is_ssl).
max_concurrency = 10 
max_concurrency_per_host = 3 
async def main(): 
	connector = aiohttp.TCPConnector(limit=max_concurrency, limit_per_host=max_concurrency_per_host) 
	async with aiohttp.ClientSession(connector=connector) as session: 
		# ... 

Also easy to implement and maintain! Here's the output with max concurrency set to three per host.

time python script.py 
real 0m16,188s 
user 0m1,311s 
sys 0m0,065s

The advantage over Semaphore is the option to limit the total amount of concurrent calls and requests per domain. We could use the same session to scrape different sites, and each one of those would have its own limit.

The downside is that it looks a bit slower. Run some tests with more pages and actual data for a real-case scenario.


You now know that scraping is I/O-bound. But what if you needed to mix it with some CPU-intensive computations? To test that case, we'll use a function that will count_a_lot (to one hundred million) after each scraped page. It's a simple (and silly) way to force a CPU to be busy for some time.

def count_a_lot(): 
	count_to = 100_000_000 
	counter = 0 
	while counter < count_to: 
		counter = counter + 1 
async def extract_details(page, session): 
	async with session.get(f"{base_url}/{page}/") as response: 
		# ... 
		return product_list

Run the asyncio version as before. It might take a long time ⏳.

time python script.py 
real 2m37,827s 
user 2m35,586s 
sys 0m0,244s

Now, brace for the hard part. Don't worry, we'll guide you through every step.

Adding multiprocessing is a bit challenging. We need to create a ProcessPoolExecutor, which "uses a pool of processes to execute calls asynchronously." It'll handle the creation and control of each process in a different CPU.

What it won't do is distribute the load. We have a solution: we'll use NumPy's array_split, which will slice the pages range into equal chunks according to the number of CPUs.

The rest of the main function is similar to the asyncio version but changes some syntax to match the multiprocessing style.

The essential difference is that we can't call extract_details directly. We could, but we'll try to obtain the maximum power by mixing multiprocessing with asyncio.

from concurrent.futures import ProcessPoolExecutor 
from multiprocessing import cpu_count 
import numpy as np 
num_cores = cpu_count() # number of CPU cores 
def main(): 
	executor = ProcessPoolExecutor(max_workers=num_cores) 
	tasks = [ 
		executor.submit(asyncio_wrapper, pages_for_task) 
		for pages_for_task in np.array_split(pages, num_cores) 
	doneTasks, _ = concurrent.futures.wait(tasks) 
	results = [ 
		for item in doneTasks 

Long story short, each CPU process will have a few pages to scrape. There are 48 pages; assuming your machine has 8 of those, each process will request 6 pages.

And those six pages will run concurrently! After that, the calculations will have to wait since they are CPU-intensive. However, since we have many CPUs, they should run faster than the pure asyncio version.

async def extract_details_task(pages_for_task): 
	async with aiohttp.ClientSession() as session: 
		tasks = [ 
			extract_details(page, session) 
			for page in pages_for_task 
		list_of_lists = await asyncio.gather(*tasks) 
		return sum(list_of_lists, []) 
def asyncio_wrapper(pages_for_task): 
	return asyncio.run(extract_details_task(pages_for_task))

This ☝️ is where the magic happens! Each CPU process will start an asyncio with a subset of the pages (e.g., from one to six for the first one).

And then, each of those will call several URLs, using the already known extract_details function.

Take a moment to assimilate that. The whole process goes as follows:

  1. Create the executor.
  2. Split the pages.
  3. Start asyncio per each process.
  4. Create an aiohttp session and create the tasks of a subset of pages.
  5. Extract data for each page.
  6. Consolidate and store the results.

And here are the execution times. The user time plays a notable role here! Here's the script running only asyncio:

time python script.py 
real 2m37,827s 
user 2m35,586s 
sys 0m0,244s

And here's the version with asyncio and multiple processes:

time python script.py 
real 0m38,048s 
user 3m3,147s 
sys 0m0,532s

Did you spot the difference? The first one took more than two minutes, while the second took just 40 seconds. But in total CPU time (user time), the latter was over three minutes! The difference is due to system overhead and all that.

That shows that the parallel processing "wastes" more time (in total) but finishes before. It's up to you to decide which method to choose. Take also into account that it's more complicated to develop and debug 😅.


We've seen that asyncio might be enough for scraping since most of the running time goes to networking, which is I/O-bound and works well with concurrent processing in a single core.

That situation changes if the gathered data requires some CPU-intensive work. We showed you a silly example with counting, but you get the point.

In most cases, asyncio with aiohttp (which is better suited than requests for async work) gets the job done. Add a custom connector to limit the number of requests per domain and total concurrent ones. With those three, you can start building a scraper that can scale!

It's also important to allow new URLs/tasks (something like a queue), but that's for another day. Stay tuned!

Or consider ZenRows if you want to save time and resources. Its powerful algorithm allows you to crawl large amounts of data in minutes and bypass all anti-scraping protections your scraper encounters. Try it for free today!

Ready to get started?

Up to 1,000 URLs for free are waiting for you