Overview
Collibra is an industry-leading governance platform for both data and AI that ensures customers have confidence in their data. Large enterprises around the globe from financial services, to retail and healthcare use Collibra to enable compliance and data quality for their AI workflows and data-driven decision making. Originally an on-premises solution, Collibra eventually turned to the cloud to pick up the pace of innovation and continue delivering best-in-class services to their customers.
Evolving cost optimization strategies from the data center to the cloud
At the beginning of their cloud journey, Collibra’s top priority was maintaining availability and security for their customers, while treating cloud costs the same way they did on-premises for the time being. Feature development accelerated and Collibra began experiencing rapid growth. One cloud eventually became two, and bills began to pile up due to high on-demand usage with only 25-40% commitment coverage in AWS and very little to no coverage in Google Cloud. The team at Collibra realized that their current operations needed more cost-efficiency in order to scale further.
“Our coverage initially was really slim. We weren’t sure what (usage) optimizations were coming down in the pipeline and we had less familiarity with GCP regarding how we were going to utilize the infrastructure at the time. We didn’t want to make a lofty commitment only to have to right-size 25% of the infrastructure and be over-committed.”
Collibra realized the complexity of managing costs for their multi-cloud estate as they started to transition some VM-based workloads to managed database services in both AWS & Google Cloud. They were initially unsure whether or not to apply commitments — often holding back due to fear of being financially locked-in. Differences in billing schema, rulesets, and capabilities between discount instruments across AWS and Google Cloud added more complication, which could leave them susceptible to wasted spend and missed savings opportunities with manual management. Collibra sought a strategic partner to help optimize costs for this new multi-cloud environment. They chose ProsperOps to automate their commitment portfolio for both their AWS and Google Cloud deployments.
“We realized the devil’s in the details and we wanted a strategic partner that understands the nuances, especially in a multi-cloud environment where we can’t apply the same AWS principles to GCP. The way those instruments work are not the same.”
World-class savings on autopilot enabled more sophisticated FinOps strategies
ProsperOps automatically doubled Collibra’s effective savings rate (ESR) in AWS from 17.8% to 34.5% in the first year following adoption. Today, over 96% of spend in AWS and 85% of spend in GCP are covered by commitments, achieving 41% and 42.9% ESR respectively. Their industry-leading savings rate was achieved with no on-going effort or additional overhead. Collibra now focuses on advancing their workload optimization strategies like building bespoke tooling to dynamically rightsize disks and running proactive spend analyses like projecting the cost of a new product before it goes to market.
Collibra no longer has to think twice about how workload optimizations will affect their commitments. Advanced risk mitigation strategies let Collibra maintain aggressive discount coverage across both clouds without the risk of being locked-in to using any one resource or service. Product and engineering teams have the flexibility to rightsize instances, spin down storage volumes, upgrade machine types, and move workloads across regions at any time without the financial loss of abandoning commitments before expiration.
“We didn’t want commitments to prevent us from cleaning up waste. That lower commitment lock-in risk lets us optimize usage without worrying about what we’d be losing because the commitments can adjust accordingly.”
Collibra’s commitment portfolio is now reaching a FinOps “nirvana” state where they are both maximally covered and have the technical flexibility to adjust their coverage at any time if usage or business needs were to suddenly change.
“We’ve built that flexibility to where we can optimize (usage) and not even think about our commitments. We can keep reducing, we can keep flipping, month over month, day over day, those reserved instances and savings plans that we laddered to operate efficiently.”
Capturing cloud savings to reinvest for innovation
Collibra’s world-class cloud efficiency lets them scale operations and services without financial friction. With commitments on autopilot and proactive workload optimizations in place, Collibra is able to reinvest what would have been spent on hosting costs into building out infrastructure for new product development and AI initiatives.
“The thing that keeps me up at night, especially nowadays, is AI consumption. But, the reason I can turn my attention there is because ProsperOps is taking care of the things that used to keep me up at night.”
Rock-solid FinOps also enables Collibra to take a more proactive approach with their overall fleet management, anticipating customers’ future needs and being prepared to build and innovate using the cloud resources that make the most sense. Collibra’s fully mature multi-cloud commitment portfolio gives them more freedom to choose the cloud provider and services that best suit their workloads without the financial constraints of long-term commitments.
“Having ProsperOps allows us to be on the offensive when it comes to cloud services and resource decisions. For example, if one cloud is offering AI services that are a better fit for the use cases we need, we have the option and confidence to double down there.”
Summary
Collibra’s industry-leading unified data and AI governance platform is supported by a solid foundation of cost-efficient cloud operations. Through a combination of intelligent automation with ProsperOps and sophisticated workload optimization strategies, they’re extracting maximum business value from their multi-cloud investments and re-deploy savings to building the innovations that their customers need to get the most out of their data.
