Workday Rides the Metrics Wave, Cruises to Wavefront’s Customer Excellence Award

“Monitoring is a really probably an overused term in the industry. Monitoring for us actually means four things. We say monitoring, but it’s not strictly monitoring. For Workday, it means alerting, stats, metrics, and logs. That’s monitoring for us. Here’s the way I explain this to people when they say, “Hey, I just want one tool that tells me all the things.” It’s kind of difficult to have one tool because you have an entirely different set of data that you need to present for different pieces. I think of it like it’s a house and the house contains all the data, but you’ve got different windows into the house that you need to look through to get different views. You’ve got the logging window, and you’ve got the metrics window, and you’ve got the stats, and then you’ve got alerting.”

“About four years ago, we decided we needed to heavily invest in the whole monitoring area. We decided that this was critical to Workday. We were doing some, the alerting, before then. We ‘kind of’ knew whether something bad was happening. But we needed to invest more in this area.”

“At the time, monitoring or alerting used to be part of every team, but then we took it out and we made it a team of its own. We said, ‘Okay, you are the monitoring team. Your job is to add all the tools that we need to monitor our services and to make them really, really great.’ We then decided to spin off this team. You can’t fix all the problems at once, so what we did was we decided to create a roadmap where we fixed specific things. The first thing that we set our sights on was alerting. We wanted to be better at knowing if something is going to fail before our customers know. We decided to put a lot of effort into alerting and improve how we alert. Now we generally know if something is happening before our customer knows it.”

“Then we decided to invest in logs so we decided, ‘Well, we want logs because we want the extra data.’ We are going to push off metrics, but then we became aware of a new type of monitoring tool built around metrics, at a new company, Wavefront.”“The value that gave us, which was really quite important at the time, was we didn’t have a lot of effort to spare. We couldn’t spend a lot of engineering cycles making a metrics platform work. We kind of wanted something that was effort-free. In fairness, Wavefront gave us something that was effort-free. It gave us metrics with very little set up on our part. We were able to just pump our metrics data through the system and people got access to that. We were really happy because we checked off the metrics checkbox on our to-do list, and with very little effort on our part and it meant that we could concentrate on other areas that were also important to us.”

“Three things that we find really great about working with Wavefront, the first is support. When we have an issue, we generally get someone who’s extremely knowledgeable very quickly and they resolve our issue generally in a very short amount of time. That’s gold. That really is gold. In case you’re thinking if that’s really a big deal, well, that doesn’t always happen with other vendors.”

“Second, we have never noticed an operational blip. That’s really valuable because we use your tool to know when our customers might be suffering. If your tool is down, then we might not know that our customers are suffering. By you never have operational blips, it means that our customers get a better service from us.

“The third thing is the product works. It does what it does, metrics. It allows us to put metrics in, see them, and use analytics to see more. That’s important to us. Sure, it doesn’t solve the whole monitoring problem, but I don’t think you’re trying to solve the whole problem either, and no one tool can solve it all.”