OpenTable selects AppDynamics to monitor system performance

OpenTable is the world's leading provider of online restaurant reservations, seating more than 14 million diners per month via online bookings across approximately 31,000 restaurants throughout the United States, as well as in Canada, Germany, Japan, Mexico and the United Kingdom. In addition to the company's website and mobile applications, OpenTable powers online reservations for nearly 600 partners, including many of the Internet's most popular global and local brands.

Challenge: Limited visibility into production

Alan Novitskiy, Performance Lead at OpenTable, was responsible for ensuring high performance and throughput of OpenTable applications. “I work to detect bottlenecks, report them to engineers and make sure that we optimize the sections of our applications that are responsible for a good dining experience,” he said.

“Our applications really need to be fast because our major competitor is the phone,” Novitskiy said. Because online reservations should be a more convenient experience, OpenTable ran the risk of losing business if diners had to revert to phone reservations due to application performance problems.

However, even with rigorous pre-production testing, Novitskiy and his team could not always predict problems that would arise in the production environment. “We performed thorough code analysis and rigorous benchmarking in pre-production environments, but when we released code into the wild all kinds of new variables were introduced, and we didn't have full visibility into what those different variables were,” said Novitskiy.

Powerful new insight into application performance

OpenTable searched for an application performance management (APM) solution that could provide the increased visibility they needed to maintain top performance. AppDynamics was one of several solutions they evaluated.

“We put AppDynamics through our benchmarking environment to make sure that our overhead was within promised numbers, and we found that to be acceptable,” he said. “So we moved further- we installed it on a single server, on a single node in our production environment, and actually started looking at what kind of data it would produce for us.”

In the first few months of having it running in our production environment we already have a much better and clearer picture of our systems.

Novitskiy was impressed with how the AppDynamics solution detected issues automatically. “It's really able to tell you about problems before you even know they exist,” he said. “With other solutions, you have to point them specifically at areas that you're interested in monitoring.” He added, “AppDynamics is really good about just going out there and detecting key areas of your application and starting to collect statistics about them.”

Another differentiator was the solution's ability to take performance averages and flag potential outliers without predefined thresholds. “I think that's really important because it's much better for detection than having a set number,” said Novitskiy. “With other solutions, what I've seen is you set thresholds and static numbers, for example, the number of milliseconds that this transaction is supposed to take.”

Comprehensive view of system performance

Within the first few months of deployment within the OpenTable production environment, the AppDynamics solution provided the additional insight Novitskiy was seeking. “We already have a much better and clearer picture of our systems and how everything interacts, and we've spotted a few specific issues that we were able to address pretty quickly,” he said.

I really liked how the AppDynamics solution detected issues automatically - it's really able to tell you about problems before you even know they exist.

“I think you should let your application monitoring solution tell you about problems that you haven't thought about. We often tend to focus on those key business transactions that directly contribute to the bottom line and sometimes we fail to see the other things in the system that could be affecting performance,” remarked Novitskiy. “If your solution is able to give you a better, clearer picture of your whole system, that's really powerful knowledge.”