What is most valuable?
- Ease of installation
- Ease of use
- It provides a common platform for Dev and Ops to be able to pinpoint and communicate problems
- The solution allows us to quickly pinpoint problems
- There’s good integration between all the products in the suite – browser and some of the plugins
How has it helped my organization?
Being able to quickly figure out the root causes of issues. It also makes it very easy to share that information with developers and other people within engineering – we can drill down in parallel.
What needs improvement?
There are a number of plugins that New Relic makes. It would be nice to be able to instantly integrate that with APM. Right now, they’re in their own little area, so it’s not as easy to quickly dive into a problem, for example in PHP. It’s a little hard to get data on the back end.
What do I think about the stability of the solution?
The stability and uptime have been great. They do a great job in notifying of maintenance windows. We have had some graph inconsistencies which might represent a quirk in the solution, but we haven’t seen that in a while.
What do I think about the scalability of the solution?
It’s a very scalable solution. We have it on maybe a couple of hundred hosts today. We use an automation tool, Puppet, so we’re able to quickly install the New Relic agent across classes of servers easily. On their end as well, they’re collecting more data, so it seems like they’re scaling out very well too. There are no performance issues with scale.
How are customer service and technical support?
Actually we haven’t had to use it yet.
Which solution did I use previously and why did I switch?
Prior to New Relic, we did not have an APM solution. We had a lot of home-built tools but nothing like this with dashboards. Mainly the maintenance of the home-built tools was difficult. It was tribal knowledge with no documentation, so we were constrained by the engineers and their specific knowledge of the tools.
How was the initial setup?
Very straightforward – it was great that everything was packaged so we didn’t have to do too much customization. We changed maybe two lines in the config. The documentation was very straightforward. Once we installed it a couple of times manually, we were able to script it, which was very easy as well.
Which other solutions did I evaluate?
There weren’t really any contenders – it was go with our home grown solution or this. Initially the product effectiveness was the top of the list. Cost is a consideration. Really cost was secondary to whether the tool actually worked and delivered value. Support wasn’t high on the list but it was a consideration. Mainly the ease of use and product effectiveness are the two main considerations.
What other advice do I have?
It’s a solid product. They continue to innovate year after year. They’re getting closer to a 10, but because of the speed of innovation I think there are a few disconnects within the suite/product line. That’s the main thing that keep sit from being a perfect solution, but they’re a very solid product and a very solid company.
It needs to be a cross-organizational evaluation. Can’t be Dev or Ops-only. The solution definitely needs to be low friction to get it into the environment. Being open to the customers in terms of what their product roadmap is and what the customer can expect, and then getting feedback from the customer to help them along as well is important.
Reviews are important. A lot of times other companies are trying to solve similar problems. People are going to be trying competitive solutions, so getting feedback is important to the vetting process.