How Fannie Mae Doubled Its Software Output With DevOps
In this article, we take a look at a case study of how a DevOps environment was able to help double a mortgage company's software output.
Join the DZone community and get the full member experience.
Join For FreeIn an article on CIO.com, Senior Writer Clint Boulton speaks with newly appointed Fannie Mae CIO Fredric Veron on how integrating DevOps has drastically improved the mortgage company’s output.
Veron has been a staunch proponent of enterprises embracing the DevOps model for almost a decade. Fannie Mae joined Walmart and Capital One as some of the larger enterprises to adopt DevOps. The mortgage provider finished revamping its programming culture 18 months ago, fully embracing DevOps, which enables Continuous Integration and application deployment. Veron claims that it has helped the company deliver software as quickly as 10 weeks compared to what was typically a six-to-nine-month process two years ago.
It is estimated that Fannie Mae has experienced productivity gains up to 40% at almost a third of the cost. Their software is also rolling 12 times faster the speed of pre-DevOps deployments and at a 30% to 48% improvement in quality, according to Veron.
Veron believes that there are rich rewards to reap when IT and business line leaders who build software together in an Agile fashion embrace DevOps. The Fannie Mae CIO says that combining Agile and DevOps allow programmers to cautiously integrate and deploy software.
“If you do Agile without DevOps, it’s like you’re trying to race with a tractor instead of a car,” joked Veron. “You can go and do the laps, but it’s not going to go very fast, you’re probably going to consume a lot of fuel, and it won’t be a lot of fun.”
At Fannie Mae, DevOps automates what used to be manual software development. The process inched along as a set of product requirements were emailed to IT from the Product Managers, causing a waiting time of several months for the application delivery.
Now, IT uses DevOps applications that help automate planning, code development, configuration, testing, and deployment. As Boulton explains:
“DevOps uses code-quality scanning tools from CAST Software to analyze software as it’s being developed and alert developers to bugs. Veron says he requires the CAST quality index must improve with each sprint before the code is pushed into production.
Veron says his team has also reduced server and database provisioning times from days to hours. Using container technology, which packages application and all its dependencies, developers move software from one computing environment to another across the company’s hybrid cloud. These changes enable IT to continually release and refine features, consistent with continuous integration and deployment practices.”
DevOps Is Faster and Less Risky
Yes, doubling output is nice, but it’s not enough to revamp its system. By continuously tweaking and deploying software using DevOps, Fannie Mae can find out what products or feature sets test well with mortgage brokers, homeowners, and other parties, as well as limit the risk of software failure.
“You get a better product because it’s better aligned with requirements and you get the value you expect from the software faster,” Veron says.
Veron remembers when IT initially pushed Agile and DevOps at the product managers to solve some of their delivery challenges at Fannie Mae almost 10 years ago.
“In 2016, we turned the corner and the business is pulling us more than ever before...What is important is the connection between IT and the business,” Veron says. “I feel that at Fannie Mae it is here to stay because the value is visible, people are embracing it, and we’re making tremendous progress.”
Published at DZone with permission of Yaniv Yehuda, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.
Comments