We are often asked this question – we are monitoring our systems and able to keep their uptime high. Isn’t that enough?
Unfortunately that only provides one side of the story. Yes, they are up and resource utilization on them may be well within the limits. But it doesn’t tell us
- How are the services running on those systems performing?
- How are the apps that use those services performing?
- How is the overall users experience for the apps/services hosted on those systems?
In fact, low resource utilization is very misleading because low utilization also happens when things are stuck waiting on external services response and nothing is processing.
Even when services’ performance is monitored via synthetic transactions we don’t get to know
- Which users were impacted?
- What is the response time for every request to the app or services?
- What caused the service to fail?
- What code/method caused the slowness/failure?
- Is the issue happening at database, app server or external services level?
End to end visibility is a must to catch problems early on, and be able to take timely action and avoid outages. Say for example an external service that is being called to process credit card transactions is not performing well, then the caller server will have to just wait on process credit card transaction and its resource utilization would be low. But the end users will feel the pain of this slowness and may even abandon the carts. It will not only be important to know why the issue is occurring but also it will also be valuable to know what users are impacted. In this situation with only systems monitoring things will look very good because systems are up and resource utilization is low 🙂
APPLICARE comes to rescue
Applicare provides end2end monitoring and answers the questions that are unanswered with just systems monitoring.
Applicare allows you to
- see the response time of every incoming request
- see the detailed drill down into performance of every request including requests that span multiple servers
- see the exact component or method that caused the slowness or errors
- see what resource became an issue on what system and how it impacted the services performance and finally users experience.
Applicare cuts down the MTTR significantly by providing the visibility into every method, every component, every external service called and deep dive into systems performance & health.
Although server monitoring is an important part of monitoring but we shouldn’t consider ‘mission complete’ with just servers monitoring in the environment. Deeper dive available with Applicare is a must for any business critical application.