Why Are My Proxies Slow? (Everything You Need To Know About Latency)

Proxies are useful in many ways: they can help circumvent geo-restrictions, conceal your identity, and gather data via web scraping. But if you’re researching proxies for personal or business use, you may have heard that they slow down your connection, which isn’t necessarily true.

Here, we’ll discuss why proxies might slow down your connection and break down the factors that can lead to that lag.

What Are Latency and Throughput?

Two factors affect any internet connection speed, with or without a proxy: latency and throughput. You may have heard these terms before but might be confused about what exactly they mean. We’ll start with latency.

Latency is the time between taking action and seeing the result. For example, if you press a key on your keyboard with a word processor open, you see the letter right away. The latency between pressing that key and typing the letter is almost nonexistent.

However, latency can increase with distance. The farther a web page request has to travel, for example, the longer it takes to get to the server and back to your computer to display the page. If you’re in the United States and trying to load a page from a server in Japan, the latency will be greater, and loading will take longer.

Throughput is the amount of data you can send in a single interaction. In other words, it’s how much information can be in a single request. This is not to be confused with the amount of data your computer can send — it’s the channel the data travels through. If you try to send a mountain of data through a tiny channel, it’s going to be slow. Throughput is also referred to as bandwidth.

Think of a traffic bottleneck, for example. On an eight-lane highway, there’s plenty of room for lots of traffic to move freely. You could say it has a high throughput. If, however, an accident were to close all lanes but one, traffic would slow to a near halt — the throughput would drastically decrease.

How Latency and Throughput Affect Your Connection

Now that you know more about latency and throughput, you can probably see how they affect your internet connection. Both of these factors determine the speed of your connection, often at the same time. Depending on the type of connection you’re using, you can have one of four possible outcomes:

• High Latency, Low Throughput: The slowest possible connection. Not only do you have high latency, but you’re only getting a tiny sliver of data through the channel. It might take a few seconds for a web page to display anything with a connection like this. The page styles will load near the end, and the page you’re visiting will often reorganize as different elements are finally retrieved from the CSS style sheet.
• Low Latency, Low Throughput: Not as bad as the above, but your connection will be slow because not much data can get through the low throughput channel, even with little latency. The web page will load slowly, with elements like images, banners, and headings trickling in. It often takes a while before you can begin navigating and using whatever site you’re trying to visit.
• High Latency, High Throughput: Still a little slower than what you’d want, but when a page does load, you see most of the elements come through at once. This connection takes a little longer to send the data, but more of it can go through. Headers and text blocks will often load first and then styled elements will fall into place once the CSS stylesheet comes through.
• Low Latency, High Throughput: As you’ve probably guessed by now, this is the optimal connection. Almost no time between sending a page request and receiving data, and a lot of that data comes through at once. The result is that pages load lightning-fast and are ready to be used almost immediately.

If you need a real-world example to illustrate the connection between latency and throughput, consider the difference between paying with change vs. a credit card.

Say you’re behind someone in the grocery store, and their bill is five dollars. They start paying in pennies, dropping them one by one into the cashier’s hand. It would take a very long time to get five dollars at that rate. Rather than wait, the cashier calls their coworker from the back, who uses their credit card to pay the five dollars almost instantly.

Here, the woman with the pennies is an example of “low latency, low throughput.” She could pay right away, but the rate was very slow. The coworker from the back is an example of a “high latency, high throughput” situation. It took a minute for them to get to the register, but they could complete the payment almost instantly.

What Affects Latency and Throughput?

The four combinations outlined above can change slightly because latency and throughput aren’t fixed properties. Consider your connection’s upload speed, for example, usually measured in megabytes per second (MBps). One day it might be 25 MBps and 38 MBps the next. Your general speed will remain the same, but the value changes slightly.

But what causes these changes? It differs depending on whether the variable is latency or throughput.

Latency is affected by the distance of the client computer from the server that’s sending it data. If the client’s computer is in another country, the lag time will usually be more significant. Your lag time will also increase if the server has to process a lot of data for your request — an extensive records search, for example, might slow you down if the host server can’t process a lot of data quickly.

Throughput is affected by how many people are using the network. If many computers or devices are siphoning the total bandwidth, the amount you can use decreases. This doesn’t just mean your home network. It’s your network, the server’s network, and everywhere in between. If there’s congestion on the network anywhere information has to travel before it reaches you, it can slow the connection. Going back to the traffic example, if an eight-lane highway is wide open, you can travel it easily. But if many other people are driving on it, everything slows down.

What Does This Have To Do With Proxies?

If you have even basic knowledge of proxies, you know they use servers in different locations and are a buffer between your computer and the internet. Because data has to hop around more before it gets to your computer, proxies can add latency to the connection. This is especially true when using a proxy located in a country that’s very distant from your computer or device.

What proxies won’t do is decrease your throughput. As long as the proxy server (if you’re using one) has sufficient bandwidth to handle the incoming requests, you won’t see a decrease in the bandwidth available for you to use. This is more a concern with data center proxies than residential proxies since not many people will be using a residential proxy at once. No matter which proxy you use, higher-end private proxies (like the ones we offer at Rayobyte) will have less risk of overcrowding.

The nature of your work will likely determine which type of proxy you use and how much latency you can tolerate. Residential proxies may be a little slower but won’t get banned as easily as proxies from a data center. You sacrifice a little speed for the authority of a residential IP address, which allows a higher work volume since you can use a pool of rotating proxies.

If you’re managing a highly automated account where stability and speed are both important, ISP proxies pair the anonymity of residential proxies with the ISP addresses of internet providers like Verizon or Comcast to add authority. ISP proxies are useful in any situation where you need both a high degree of anonymity and speed.

If you don’t need residential proxies to scrape the site you’re working on — you don’t have to rotate addresses to scrape thousands of pages, for example — then data center proxies will fulfill your needs just fine.

Some measures like content delivery networks (CDN) can reduce latency from proxies. A CDN is a group of servers placed in strategic geographic locations to decrease latency. They put the server closer to the end-user, so the data hops around less.

Final Thoughts

Even if your proxies do result in latency, chances are you won’t notice it much. The lag might be a few seconds at the most, and the results you get from using proxies far outweigh any inconvenience you may see from a slower connection. If slower speeds are your concern, Rayobyte offers quality data center proxies that can handle all your requests.

The information contained within this article, including information posted by official staff, guest-submitted material, message board postings, or other third-party material is presented solely for the purposes of education and furtherance of the knowledge of the reader. All trademarks used in this publication are hereby acknowledged as the property of their respective owners.