Single vs multiple Apogee sites
Apogee is Breakpoint’s Digital Experience Monitoring service. One of the many ways Apogee can test the digital end-user experience is through running transactions against digital services, over any access channel – fixed or mobile. Let’s take an example use case: monitoring a mobile service over GSM.
The way an Apogee remote simulates a user on the GSM network is by means of programmatically controlled GSM terminals with provisioned sim cards, running transaction requests every few minutes. This way, synthetic transactions are repeatedly run against the service, just like a real user would, while the responses that the service provides are measured and checked for validity against defined criteria, and then analysed, visualised and alerted on when things go outside of acceptable parameters. While a single Apogee channel which is constantly testing a network or service can be a good indication of detecting issues, it may still be unclear what the root cause of the issue is, as it may not be an accurate representation of the entire network – it simulates a single, static user that does the same transaction over and over. What about users in other parts of the network? Or users that do something slightly different? Does it affect mobile app users on smart phones, USSD users on feature phones, or both?
That’s where Apogee configured in a multi-site monitoring environment can really help – by giving a broad overview of the entire network or system, it can help with pinpointing the source of issues quickly.
The bigger picture
Mobile devices communicate to the network (and in some cases to other devices) by converting voice, text or multimedia messages into Radio Frequency (RF) communication. A cell site or base station with a cell tower receives these RF signals and connect devices to other devices and networks. A GSM mobile network consists of a number of these cell sites across the country that communicate with each other. The working range of a cell site (the range within which mobile devices connect reliably to the cell site) is affected by many things, such as for example the antennae coverage of the cell site, the reflection and absorption of radio waves by buildings and vegetation, as well as how busy the particular cell is, which is a function of how many connected devices it is carrying and how much traffic those devices generate on the network. All of these factors combine to make any wireless network slightly unreliable – it is inevitable that customers might experience dropped calls, interrupted sessions and failed transactions when operating on any wireless technology. Understanding and accepting this fact makes it plain to see why monitoring from a single channel or location is insufficient to obtain a true reflection of service quality or state.
Why having multiple Apogee remotes is more reliable
Having multiple Apogee remotes installed and connected to the mobile network at different geographical sites (thereby connecting to multiple cell sites), simulates end users across the network, creating a more reliable picture of the service state. So, while a single Apogee remote can detect errors, if that single remote was connecting to a cell site that is experiencing technical issues or downtime, it might cause you to believe that the entire network is experiencing issues, whereas that might not be the case. Having multiple remotes on the other hand, will immediately identify that the other sites are not experiencing issues, and will clarify that it is only a regional issue. Conversely, having multiple channels monitoring the same service gives you a confident result when both channels are indicating the same service issue.
What if my monitoring says that the service is healthy, but users are unable to transact?
This is key to why Apogee’s outside-in monitoring approach is the only true reflection of the real service state. Even though the service might be perfectly healthy, there may be carrier or gateway issues, or even middleware problems that prevent users from successfully transacting against your service. The final piece of the puzzle is to monitor the service directly at the interface to the network while bypassing any mobile networks or middleware stacks, and to inject synthetic transactions to the API, right at the point of handover. This is a sure way to monitor the service health, end to end, and to give you confidence in any conclusions drawn from the digital experience health indicators.