The public utility has jumped on the machine data bandwagon to maximize internal productivity.
Denver Water has harnessed the power of machine data to find potential system errors and fix them before users even realize something is amiss.
For the past nine months, the public water utility has been combing through activity logs of its internal applications looking for trends that indicate problems are on the horizon. Log data is displayed in various graphical formats as opposed to an unstructured jumble of text and code, enabling IT leaders to make more informed and proactive decisions.
Denver Water is using a data solution from Splunk, an IT operational intelligence provider.
Instead of being reactive to technical errors and help desk tickets, the utility’s IT team now can quickly see failure trends and address them ahead of time. In addition to troubleshooting, machine data is also helping the utility get a better picture of how their applications are being used by employees.
For example, the utility has an internal program called E-Map that provides extensive GIS information to users. By using machine data, Denver Water can now pinpoint the volume of people using it and what functions are being used the most, so it can prioritize upgrades to the application.
Jonathan Spitze, SCRUM team lead for Denver Water's Geospatial Asset Management team, said machine data gives him the ability to very quickly grab and synthesize trends. As an added bonus, the data is presented in a way that it can be shared with and understood by nontechnical colleagues.
According to Henri van den Bulk, enterprise architect with Denver Water, gauging whether machine data analysis is having a positive impact is easier said than done. He explained that the ultimate goal is to have higher internal customer satisfaction across the board by increasing the stability of all of the utility’s IT systems.
And while van den Bulk felt Denver Water’s IT team had improved in that area, he said there’s no quantifiable number to point to in order to illustrate success. He said the primary measuring stick for effectiveness is the fact that IT can have better conversations with users.
Spitze agreed. He said the two main things he looks at are whether IT is improving the ability of its internal customer base to use applications when they need to and whether downtime is reduced when something goes wrong.
“What I want to see is a downward trend,” Spitze said. “That I’m getting to a point where our users don’t experience outages ... errors or failures, that they’re having a good experience every time they get into these applications.”
Looking ahead, Denver Water plans to expand its use of machine data. In addition to how it’s applied in an IT setting, the utility is considering blending machine and business transaction data. So if a business event is happening, machine data may help provide more insight into an issue.
“I think we’re in the day and age now that we can start making sense of large volumes of data and it becomes useful,” van den Bulk said. “Data becomes a very powerful thing to help plan and make changes. Don’t be afraid of it, embrace it.”