Case Studies

Care.com Chooses AppDynamics to Monitor Critical Production Applications
Watch the Video

Industry: Services

Customer Since: 2011

Favorite Feature: Memory Leak Detection

Video Transcript

Speakers: Jacob Marcus, Senior Director of Engineering at Care.com and Praveen Subbarao, Senior Quality Assurance Manager at Care.com

Jacob: At Care.com we have a goal of being the premier product in our space, which is helping people find care – senior care, babysitters, nannies, etc. My name is Jacob Marcus, I’m the Senior Director of Engineering at Care.com. I take care of the development team at Care.com, and I’m also very involved in the infrastructure and architecture of how our software is written and maintained.

Jacob: We had on-and-off memory-related issues with the garbage collection. It wasn’t completely understood what was holding onto memory at one point in time, but since we have adopted AppDynamics it has really helped us nail a few issues that we’ve been chasing for a long time.

Praveen: My name is Praveen, I work for Care.com. I joined the company in 2007, I’ve been with the company for three years now, and I’ve built a QA-strong team. In terms of collaboration, the Operations, Development and QA teams meet on a daily basis, and the issues that are raised out of AppDynamics are discussed and utilized.

Jacob: We’ve started using it more in our software development cycles. Even before our software is pushed to a staging environment, we use AppDynamics to figure out exactly how many queries are getting called. The developers have a better understanding of how everything is working these days, and it gives them an opportunity to tune some of the code that they write. It’s helping the quality of the software that we write, to some extent. It also helps us to understand our peak volumes by giving us good data like number of calls per minute that you can monitor over time. You can even decide to beef up your production infrastructure to meet your peak volumes.

Praveen: Before acquiring AppDynamics we were primarily relying on open source tools like Hyperic, Nagios, Cacti, and we were traversing through logs to find errors. Using AppDynamics has reduced our mean time to achieving the root cause of the issue. The good thing about AppDynamics is getting the facts to the developer in the form of issues. Utilizing AppDynamics to get to the root cause of the issue in terms of the lines of code which brings out the issue to the developer is far more reaching to the developer to address the issue than just saying it’s broken.

Jacob: One of the concerns we had when evaluating products of this nature is, is there a delta in terms of response time for our HTTP requests? And we’ve been very careful about that, and we’ve noted that AppDynamics does not add a lot of overhead to HTTP requests at all. If it adds something, I’m pretty confident it’s negligible, because we’ve never had any issues with that.

Jacob: The amount of things that it brings with it, the ability to troubleshoot issues very quickly, have convinced us that it’s something we want to keep running in production. It just helps us a lot that way.