As the very first interaction between an end user and a digital property, DNS resolution is arguably the single most important component of any web transaction. In this sense, fast DNS is just as important as fast content, and something that every single company must take into consideration when establishing their online presence. 

It also means that DNS providers are essentially responsible for the very first interaction between consumers and businesses; this is a  responsibility that is not to be taken lightly if a vendor wants to remain in business. Given that DNS resolution can account for up to 30% of page load time, slow resolution can torpedo the user experience right from the start. 

It’s with this in mind that Oracle+Dyn, which powers the digital infrastructure of over 3,500 companies through one of the largest DNS networks in the world, takes great pains to ensure that their service is operating at optimal efficiency 24 hours a day. The last thing any service provider wants is to be the cause of their clients’ performance problems, so they must run frequent and regular tests of their infrastructure on both internal and external monitoring platforms. 

However, due to the complex nature of DNS resolution, running health checks and analysis is not as simple as a quick page load or availability test. Providers generally use multiple name servers and put other redundancies in place in case a single server or component fails, which provides necessary insurance, but also complicates the monitoring process with extra layers that need to be checked on a regular basis. 

Additionally, performance issues such as an absence of glue records, cache poisoning, insecure zone transfer, etc. can also lead to errors that require a powerful external monitoring solution which can accurately replicate the end user experience. 

A monitoring platform should be able to provide full visibility through the entire name process, but also ensure that A records and CNAE records are used to map domain names to other domain names, such as when aliases are used. It’s also necessary to test DNS resolvers, aka the client side of DNS. A resolver is a local server that stores a central database of DNS name servers and manages DNS requests for all the clients on a user’s network, whether a private corporate network or local ISP. 

Therefore, Oracle+Dyn has extensive usage of the two different types of DNS monitors that Catchpoint offers. Direct Name Server tests give them indicators of availability, which DNS Experience tests are able to provide more granular performance data. For simple up-or-down indicators (i.e. is a particular server reachable from a certain region at any given time?), the direct name server tests are sufficient. 

However, end users don’t query authoritative name servers; recursive resolvers and monitoring probes do. This is where the Oracle+Dyn network operations team heavily uses the DNS Experience tests, which leverage these recursive resolvers to measure latency from a perspective that better emulates the experience of the actual end users. 

Just as importantly, these tools are capable of being utilized within one cohesive platform, enabling the Oracle+Dyn NOC to access their DNS performance data quickly and easily without having to flip back and forth between different tools and windows. As one can imagine, when a performance issue does arise, the ability to analyze and dissect the data in a timely manner is of paramount importance. With these tools all located in one platform within the Catchpoint portal, the NOC can rest easy (as much as any operations pro can ever rest!) knowing that they have the necessary tools to identify and resolve any performance issues quickly, thus keeping their customers and end users happy. 

To learn more about Catchpoint’s DNS monitoring capabilities, sign up for a free trial.

Author Craig Lowell

Leave a Reply

Your email address will not be published. Required fields are marked *