Managing Proxy Speed

Our article Measuring Proxy Speed describes how much proxy speed contributes to the benefits of proxy services. It explains how a proxy server speeds up your internet transactions and discusses objective measures of proxy speed, with links to some speed testing services. 

Manage Latency

Latency is delay in connection and transmission of data between two computers located far apart and communicating over the Internet. The delay is generally due to the geographical distance and the number of “hops” between connecting servers.

Proxy and Target in the Same Country

You can reduce latency by choosing a proxy as close as possible to your target server. ProxyMesh provides proxy servers in many different locations. If you do use proxies in different continents from your target servers, you may encounter much higher latency.

Try the World Proxy

The World Proxy has outgoing IPs located all around the world. The World Proxy server is physically located in the US, so if an outgoing IP is in your target country, then requests through the World Proxy will likely take at least 1 second longer than a direct request. But the World Proxy offers advantages. 

Along with this access, ProxyMesh's World Proxy provides a custom header to target requests to some 22 countries.

Try the Open Proxy

Open proxies are free and available to all Internet users, and can forward requests from and to any site. Like other types of proxies, an open proxy offers online anonymity and privacy through concealment of IP address from web servers, since the server requests appear to originate from the proxy server. 

With an open proxy, however, this anonymity is not total. Also, open proxies are slower and more error prone than commercial proxies. But the tradeoff is a huge increase in quantity and diversity of IP addresses.

Check Geolocation

You can speed connectivity by making sure your location and the remote site location are geographically close. 

But keep in mind that server locations may change. Some hosting providers may reassign blocks of IP addresses to a data center in a different geolocation from their original one. And it can take some time for the geo IP databases to update the IP location so that it is accurately stated in responses to location testing.

Use Data Compression

To speed transmission and help control bandwidth usage, include an X-ProxyMesh=Accept-Encoding header to take advantage of compression options such as gzip. Most remote sites support at least one of these content encoding methods.

On the command line, you can pass a compression header with cURL, like this:

-H "Accept-Encoding: gzip"

Although the request example is for gzip, the server may apply a different compression method.

Compression may be especially useful in speeding high-traffic research with sizeable amounts of data per request. Although many proxies will strip out identifying headers, they do not alter the content of your request. So, whichever compression headers you send to the remote server will be passed through the proxy, so that the proxy will send back the requested data in compressed format. 

Minimize Requests

Minimize requests to pull images, JavaScript, and CSS files. These files usually aren’t necessary for web crawling or screen scraping.

This practice won’t speed up transmission, but you can avoid slowing it down while optimizing bandwidth use and controlling the total number of requests.

Distribute Requests

Distribute requests over many IPs to reduce delayed responses and timeouts. A rotating proxy server can help you avoid rate limits and blocking by choosing a random IP for each request.

Avoid Timeouts

Proxy servers typically have request & response timeouts. If the remote server does not respond in that time, you will get a 408 response code. If you need to wait longer for a complete response, you may be able to use the X-ProxyMesh-Timeout custom header to specify the number of seconds you want to wait.

If a significant portion of your requests are timing out (the 408 response code), here are a few possible causes:

• The network connections between you and the proxy and/or between the proxy and the remote site could be unreliable. Try some different proxies and see if that fixes the problem.
• The pages you’re requesting take a long time to load. If a proxy server normally waits for 20 seconds, you could increase that time with the <X-ProxyMesh-Timeout custom header.
• The proxy IPs have been blocked by the remote site. If you think this is the case, then you’ll want to switch proxies, and ideally use multiple proxies to distribute your requests.

Some other time-saving strategies:

• Reduce time and number of requests from the same IP address using rotating proxies, which helps prevent rate limits and blocking.
• Avoid timeouts by using proxies located near your target sites. Try configuring a custom request header for this result when using a world proxy.
• Anticipate 301 responses (i.e., a site has been permanently moved) by scripting your request to follow redirection.

Other Good Practices

Here are more recommended practices that can speed your proxy responses and minimize timeouts.

• Reduce the number of concurrent requests from a single IP. This could involve using an additional IP for crawling, or slowing down your crawl rate on your current requests.
• With added proxies, you have more connection strategies available, such as putting all of your authorized proxies in a list in your code or script, then randomly choosing one proxy for each request.

Speed is essential in a proxy operation. Measuring proxy speed can be complex. But you can find reliable services to help you understand and measure proxy speed. Try using the strategies we’ve outlined here to maximize the benefits of proxying.

For more about how you can maximize transmission speed, see our blog To the Swift: Proxies on Your Relay Team, Part Two.