New Relic APM Review

When there’s a deployment, it shows by graphing if a regression has happened, and it allows us to react. However, every now and then there are little quirks.


Valuable Features

It helps you to define the transaction percentages, average time, and highest throughput. Also, it tells us the transactions that take the most time on average. Those are the high level, most useful features.

It also tells us about every single request that comes in and how the system reacts to it. You get to see everything from the dashboard, all these breakdowns per layer of your architecture.

Error rate is the second most useful feature – there are alerts tied to that. You get paged when the error rate is above an expected percentage and that has worked very consistently and reliably for us.

Improvements to My Organization

The best thing is that the team has grown, and a lot of people are developing the code, but you tend to have regressions that are clearly visible in those transaction traces.

When there’s a deployment, it shows by graphing if a regression has happened, and it allows us to react. Catching regressions in performance is very important, and since we now see the breakdown in every single layer in the application, you know right away if there’s something you’re not expecting. We can then go and figure out if it’s an infrastructure or code issue. It gives you a high level view of all of the requests coming in. Error rates are a good indicator for potential rollbacks for a potential deployment – and usually it’s pretty instantaneous. At the end of the day as users, we get what we want.

Room for Improvement

For the purposes for which we’re using it, it just works. So far I don’t have any requests for new features.

Currently, it is not the only solution we have for monitoring so there are things that it’s missing – for example what Datadog does for us. Timeline series, custom timelines and graphs, and I’m not aware of those features in New Relic.

Stability Issues

Every now and then there are little quirks, like the web site will stop refreshing by itself, or the graphs will show something that’s not happening. But in my experience just refreshing the graphs will fix it. But we’ve never had any downtime with New Relic.

Scalability Issues

We have scaled up as well in terms of number of hosts. There is some perceivable difference in performance when you’re looking at a graph versus number of hosts, but so far it has been fine. It’s definitely not the same looking at a single host versus many hosts.

Customer Service and Technical Support

The fact that I’ve never had to contact support by email or phone is a good thing. The online documentation has been fantastic. Everything you want is available in the documentation.

Previous Solutions

It was already in production. I did deployments in the staging environment, but not the first deployment. We will be doing the first deployment for mobile as well.

Initial Setup

It was already in production. I did deployments in the staging environment, but not the first deployment. We will be doing the first deployment for mobile as well.

Other Advice

You need to understand what’s in their stack, what technologies, what libraries, and it takes someone who has experience with those technologies to help make the decision. It also comes down to best bang for your buck, and I’d definitely recommend New Relic APM.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
Add a Comment
Guest
Sign Up with Email