Kyriba unifies data management, cuts NAS troubleshooting in half

The Challenge

Kyriba’s IT environment is both diverse and complex, so compatibility was a key requirement when the company’s DevOps leader looked for a new data management solution. In addition to supporting legacy files that facilitate millions of bank transactions per day across regions, Kyriba invests in advanced engineering work spanning public clouds (AWS and Microsoft Azure), VMware workloads on-premises, Docker containers, Kubernetes, and more. The company needed a software-defined solution that could integrate with many different file formats, platforms, and APIs, and work seamlessly with existing orchestration tool sets including Chef and Terraform.

As DevOps automated deployments and platform configurations, it also wanted to automate backups. This included using tag mechanisms, for example, enabling systems that tagged production to be backed up one way and non-production systems another way. The financial enterprise looked to automate manual disaster recovery (DR) exercises which often took a day or two to set up.

“We chose Cohesity because it was agnostic. It’s not loyal to any other vendor and the S3 capability in particular was a tremendous positive during our selection,” said Matthew Dartez, Kyriba Director of DevOps and Infrastructure.

Kyriba also wanted to more efficiently and securely support the hundreds of millions of files in many bank formats across localities constantly being written, read, and converted, and ensure the same data management policies became de facto standards for all of its code. The enterprise had already invested in Splunk for machine-learning analytics and wanted a modern NAS solution that would seamlessly integrate with it and be more than just a fast backup target.

The Solution

Kyriba chose Cohesity to help it automate operations and scale, anticipating 15–30% increases in its data footprint every year. “Everything’s automated—the way we program and build. We integrated Cohesity into all of our solutions, so as we scale out, the platform scales with us,” said Dartez.

Kyriba operates a highly regulated hybrid cloud environment, in compliance with global industry and government mandates such as GDPR, CIS, SWIFT, and SOC. The company requires robust backup and recovery capabilities, including features that enable Kyriba to meet seven- and 10-year data retention requirements. Built-in security such as encryption of data at rest and in transit are also essential.

Cohesity also streamlines compliance processes by allowing the company to customize its storage tiering as well as use S3 Glacier as a target to archive data for its data warehouse. In contrast to legacy products, Kyriba can use Cohesity to restore an entire cluster in a region of AWS or Azure, as well as another virtual machine (VM) or physical data center without an issue, in a couple of hours.

The financial company is working with Cohesity SmartFiles, gaining software-defined flexibility for enterprise file services. Kyriba will replace tiny file servers with a unified central set as a shared services target platform. For example, in many physical data centers the company will use a few different file server clusters to centralize all customer NAS instead of having hundreds.

Looking ahead, DevOps is teaming with Cohesity on plans to implement A/B testing for its continuous integration/continuous delivery (CI/CD) pipeline for better incident prevention and higher quality testing. Continually and quickly restoring the environment automatically instead of manually as it breaks without recreating data access would be another significant time saving.

The Results

By reducing IT complexity and automating with Cohesity, Kyriba is already seeing backup and recovery improvements and anticipates more in three key areas. The first is better service level agreements (SLAs) for restores. Today, the DevOps team SLA is just over four hours for production and eight hours for non-production systems. With Cohesity, the team is targeting a two-hour improvement. The second is shortening the DR testing exercise times from several days to just one day, and restoring an entire platform that used to take one to two days in just a few hours—without issue during the normal business day. The third is lower operational overhead. Kyriba’s old backup systems required considerable and continual manual intervention to continue their successful RTO/RPOs. Now the team has a very rigid workload tagging strategy where everything is automated as much as possible and checked constantly. Cohesity policies are integrated into it for 100% enablement.

When Kyriba had to restore a database with some financial transactions to fulfill a client’s audit requirements, Cohesity made it simple. Dartez said, “Typically a small task like that would have taken us about an hour and this case took us a couple of minutes.”

With Cohesity going beyond scale-out NAS, Kyriba anticipates incident reduction due to fewer legacy operating systems as well as an approximately 50% reduction in troubleshooting.