Covering the latest industry trends and an excellent source of thought leadership.
Once an internet application is built, operating it requires diligent monitoring 7 x 24 x 365. Monitoring in a broad sense means to track certain measures and when those measures reach a certain threshold to take some type of action. Traditionally for Internet Applications, monitoring means Pinging the IP address. To Ping an IP address is to send a small packet to that IP address and see how long it takes for it to come back. This tells you if your website is available to users on the Internet – i.e is the site up? But this is kind of like your doctor saying you’re fine because since he’s talking to your you must be alive. Obviously a healthy site, like a healthy person, requires knowing a little more.
The next generation of monitoring also tells you how well the site is operating and if it’s not operating withint a certain threshold to notify someone who can proactively address the measure being monitored. Kind of like taking a cholesterol measure of your body to tell how healthy you are (or aren’t). It provides a predictive measure.
There are many predictive measures you can use in your internet application. It’s important to build these predictive monitors into each layer of the Internet Delivery Stack (IDS). For example, the log in page of your website could track the number of seconds it took to log in a user and notify an engineer when it takes longer than 3 seconds – for something might be brewing and you’d like to address it before customers just stop trying to log in. This is an example of monitoring at the application level of the stack.
So, as you design, build and manage your internet application think about monitoring for predictive measures, not just reacting to notification that something is down.