Rent-A-Center detects future anomalies with Cognitive Insights

Rent-A-Center detects future anomalies with Cognitive InsightsRent-A-Center detects future anomalies with Cognitive Insights Rent-A-Center detects future anomalies with Cognitive InsightsRent-A-Center detects future anomalies with Cognitive Insights
"Cognitive Insights gave us confidence that we would be able to detect future anomalies with"
Troy Washburn
Senior DevOps Manager
Plano, TX

Rent-A-Center (RAC) is a Texas-based furniture and electronics rent-to-own company that offers new and used brand-name furniture, appliances, computers, and electronics to customers. RAC is a Fortune 1000 company operating 3,000 brick & mortar stores, 2,000 kiosks within third-party retail locations, and online rent-to-own business lines within the United States, Canada, Puerto Rico, and Mexico.

Rent-A-Center is successfully undertaking a digital transformation that will help foster innovation and deliver great customer experiences with DevOps at the core of the transformation.

Using ELK at Rent-A-Center

Rent-A-Center initially started down the ELK path as a way to centralize infrastructure (AWS) and application logs for analysis and troubleshooting and remove the necessity for developers and engineers to log directly into production systems to view logs.

This was extremely important as the company moved into a dynamic, auto-scaling environment. Rent-A-Center has since moved on to creating visualizations, dashboards, and alerts to proactively analyze system and application performance and health.

Before moving to, Rent-A-Center had created two different ELK stacks (one on-premise and one in the cloud). As the company grew its in-house solutions to ingest over 100GB per day, Rent-A-Center spent a lot of time optimizing them by dealing with additional input capabilities, memory tuning, disk space management, data archival settings, and master/data node changes.

While setting up an ELK stack was fairly straightforward, optimizing it for performance and reliability took a significant amount of ongoing attention and effort from Rent-A-Center.

Feeling the Pain

Rent-A-Center did not have a dedicated team to manage their ELK installations, so competing priorities made it a challenge to maintain their existing stacks. This led the company to have out-of-date versions of the different components in its stack as well as stability challenges because Rent-A-Center often could not devote time to the stack until after an incident had already happened.

“Cognitive Insights gave us the confidence that we will be able to detect future anomalies with”

Troy Washburn, Senior DevOps Manager

Going for an Enterprise-Grade Solution

Moving towards a SaaS solution for ELK was a no-brainer that allowed Rent-A-Center to focus its efforts on higher-priority tasks and functions that related more to its core business. utilized similar technologies in which Rent-A-Center had already invested as well as new, enterprise-level capabilities including pre-built dashboards and Cognitive Insights that provided enhancements over what Rent-A-Center had already built.

Cognitive Insights and its machine learning capabilities had already proven to be a very useful tool for Rent-A-Center. During the pilot phase, it alerted the company to a situation with multiple failed root login attempts that Rent-A-Center previously did not detect. This security use case helped the team isolate potential compromise or ongoing attacks which originate externally and internally.

After the security team reviewed the situation, it ended up being an internal penetration scan. While this was a planned scan, it gave Rent-A-Center confidence that the company will detect future anomalies with

Transitioning to

Since Rent-A-Center had originally designed its solution to have all systems and applications forward to the company’s centralized Logstash cluster, one quick first step was to split the traffic and send it to both Rent-A-Center’s on-premise ELK stack and the ELK stack. made this quick and easy by detailing the exact steps on how to configure Logstash. This allowed Rent-A-Center to start taking advantage of’s extra capabilities immediately. The company is currently in the middle of the rest of the transition, but Rent-A-Center is working methodically to convert applications to send logs directly to to avoid having the need to manage any on-site infrastructure components at all.

The team expanded data sources with to include more AWS services for both operational and security data AWS Cloudwatch, AWS CloundTrail, and AWS CloudFront logs.

There are close to 100 employees at Rent-A-Center that are transitioning from using its original on-premise ELK stack to Many of the company’s developers use it to gain visibility and transparency into how their applications are functioning in all environments up through production.

Rent-A-Center’s application support and operations teams can quickly correlate events across all systems and applications, allowing them to determine impact of any issue much faster and to proactively identify issues. The company also has multiple dashboards that the leadership team and executives use to determine high-level health levels across several of the Rent-A-Center’s applications.

In addition, because runs on AWS,  customers can easily run their full logging pipelines at scale as an enterprise-grade solution. Furthermore, AWS makes adherence to industry and government requirements relatively simple by aligning with IT security best practices and following a number of compliance standards such as HIPPA, SOC 2, and FISMA.

The Results

By turning over their ELK stacks to, Rent-A-Center increases their agility by gaining the ability to easily scale with growing volume.  In addition, with the frequent stack updates and improved stability & reliability, Rent-A-Center has even started to take for granted that their logging solution is always available and current.  Overall, has freed Rent-A-Center’s time from maintaining the company’s original on-premise ELK stacks to allow the IT team to focus on more important business functions.

Get started for free

Completely free for 14 days, no strings attached.