Blog Post

CDN Monitoring – Why You Must Monitor Your Extended Infrastructure

Although CDN services are now an indispensable part of any application infrastructure, visibility into CDN performance remains limited.

The content delivery network (CDN) has been an integral part of application infrastructure for more than two decades. A CDN is critical to the end-user experience, but it is no longer considered to be just a caching server. It has evolved to provide security from cyber threats, including DDOS attacks along with front end optimization.

Although CDN services are now an indispensable part of any application infrastructure, visibility into CDN performance remains limited. Many enterprises do not have a strong CDN monitoring strategy in place, even though most of the cached content is handled by CDNs and almost all traffic to the origin is directed through a CDN.

In this blog post, we will talk about the top four reasons to invest in CDN monitoring.

1. Detect And Mitigate Major Outages That Impact End-User Experience

CDNs rely on specific software to manage its functionality. Software updates are routinely carried out to ensure optimal service but there has been instance when such software updates have resulted in configuration errors leading to CDN downtime.

Figure 1: End users unable to access application due to issues with the CDN.

The recent Akamai and Fastly outages brought down thousands of websites and severely compromised end-user experience. Such outages have an impact not just on business revenue, but also brand reputation. The scatterplot graph below (figure 2) illustrates the impact of the Akamai outage.

Figure 2: CDN outage impact.

When handling such an outage, it is vital to have access to detailed performance data. The data will enable you to quickly implement measures to mitigate performance degradation. This minimizes the negative impact that the outage can have on company reputation and revenue.

These outages are inevitable and will occur again in the future. At the same time, the onus is on your company to ensure your application's availability, reachability, reliability, and performance. How can you achieve this seemingly impossible goal when you don’t have full control over every aspect of the situation?

To start with, you can implement a robust CDN monitoring strategy which will accomplish the following goals:

  • Notify the right teams with accurate data immediately, to avoid chaos during an outage.
  • Provide detailed data to make it easier to troubleshoot and resolve the issue. For example, identify if the outage impact is local to one location, country, or IP address.
  • Adopt a multi-CDN strategy and origin redirection to greatly mitigate the impact of a CDN outage. Note that the success of this move is dependent on also implementing a comprehensive CDN monitoring strategy.

2. Identify Minor/Micro Business-Impacting Outages

Major outages are usually widely reported and make it to the headlines. However, CDN services also suffer from smaller, region-specific outages that may not have a global impact. Such minor outages should be monitored, as they can still impact end-user experience significantly.

In the graph below (Figure 3), the CDN servers in India were having issues, while the origin performance remained consistent with 100% availability. It’s difficult to evaluate CDN performance accurately if such inconsistencies are ignored.

Figure 3: CDN performance degradation impacting a specific region.

If you are overlooking monitoring such minor outages, then there are two things to consider:

  • While CDN providers do monitor their infrastructure, they don’t monitor individual customer accounts.
  • Companies spend a lot of money third-party service vendors but have zero visibility on SLA breaches. CDNs evaluate huge volume of traffic. Individual client content might vary by just a very small percentage (varying from 1% to 20% per CDN POP), making it difficult to report an SLA breach if you aren’t already utilizing a monitoring strategy.

3. Evaluate DNS Performance of CDN

Outages can be caused by several different variables in the CDN infrastructure. For example, server reachability could have a negative impact on CDN functionality. The server maybe experiencing DNS or other network related issues.

Many companies proactively monitor DNS service providers as part of the process of managing their infrastructure. At the same time, they often don’t measure CDN services. It’s important to do so because the CDN itself comprises of different components and configurations that determine the application performance.

It is important to monitor DNS performance. The graph below (Figure 4) indicates the variation in DNS resolution of a domain's Authoritative Name Server from multiple locations. You can see that it’s only possible to accurately evaluate whether the issue comes from the CDN or Origin only by looking at the results from monitoring DNS performance, including DNS timeout and availability for the different domains.

Figure 4: Evaluating DNS performance – DNS provider vs CDN.

4. Improve Performance Globally

The primary reason for using CDN is to offload the cacheable content from Origin. That said, the objective is to improve the performance of the end users globally. It's not practical to deploy thousands of servers in multiple locations to support application delivery. Instead, the CDN network, with its caching servers, maintains performance and end user experience.

The value of CDN can be easily understood by comparing application performance with and without a CDN. The improvement should be at par with the Origin, if not better. The graph below (Figure 5) illustrates performance using CDN and performance from origin. It is obvious that in all scenarios, CDNs have better performance.

Figure 4: Evaluating DNS performance – DNS provider vs CDN.

Conclusion

The four points discussed here are just a few of the benefits that CDN monitoring adds to a company's overall monitoring strategy. CDN observability enables businesses to understand whether a high web page response time is due to objects that are either not cached or not cached for longer duration. Evaluating the cache hit/miss ratio provides a clearer understanding of performance.

Because CDNs greatly impact your global end-user experience and all data goes through the CDN infrastructure, it is very important to gain visibility into that network. When you have better CDN monitoring, you can make better decisions around key CDN related strategies such as implementing multi-CDN, evaluating cost vs offload, or picking geo-specific CDN vendors.

Ready to discover how CDN Monitoring can help improve your own business? Learn more about CDN Monitoring here!

CDN
Incident Management
This is some text inside of a div block.

You might also like

Blog post

The Future of Operations: AI-powered Internet Performance Monitoring

Blog post

The High Price of Internet Disruptions: New Study Reveals the Financial Impact on eCommerce Companies

Blog post

Preventing Outages in 2023