How Priceline Enhanced Kubernetes Reliability at Scale with Komodor

Company Size:



Internet Publishing

Komodor Installation:

18 Clusters | 500 Nodes

About Priceline, part of Booking Holdings, Inc. [NASDAQ: BKNG], is an online travel agency for finding discount rates for travel-related purchases such as airline tickets and hotel stays. The company facilitates the provision of travel services from its suppliers to its clients. It operates in more than 200 countries and territories around the world and has partnerships with over 400 airlines and 300,000 hotels. 

The Challenge

Priceline’s decision to adopt Kubernetes to enhance its digital infrastructure led to a significant increase in system complexity. This complexity, in turn, resulted in delayed remediation during operational issues. While the deployment of multiple replicas across Availability Zones ensured minimal downtime, the intricate nature of the Kubernetes environment made identifying and resolving problems more time-consuming. This delay in troubleshooting and fixing issues had a cascading effect on the software development cycle, causing slowdowns in the release of new features and updates.

The Problem

This situation impacted the customer experience, as the extended time required to address and rectify issues meant that improvements and new functionalities were not reaching customers as swiftly as needed. Therefore, simplifying and streamlining the Kubernetes management process became an essential business objective for Priceline, emphasizing the need for effective and efficient operational workflows.

The Solution

Facing a host of Kubernetes-related challenges, Priceline decided to engage with Komodor, a platform specializing in Kubernetes troubleshooting and operations.

Komodor brought unparalleled visibility into Priceline’s Kubernetes ecosystem, offering comprehensive insights that streamlined decision-making.

This clarity accelerated troubleshooting by quickly identifying issues, substantially reducing both diagnosis and resolution time. 

As a result, Recovery Time Objectives (RTO) were reduced, minimizing the impact of downtime on both revenue and customer experience.

Moreover, faster problem identification meant that Recovery Point Objectives (RPO) were also improved, reducing the risk of significant data loss during system failures.

Beyond technical fixes, Komodor also had an organizational impact.

It enabled a “shift-left” approach, empowering developers to catch and resolve issues earlier in the development lifecycle. This reduced the frequency of production errors and significantly boosted reliability across the system. 

Additionally, the tool facilitated better collaboration among Priceline’s various teams. Its easily shareable insights made it simpler for departments to sync up, thereby streamlining both communication and problem-solving efforts. 

Overall, Komodor provided Priceline with the tools they needed to overcome their Kubernetes challenges, significantly improving operational efficiency and development speed.

Download this Case Study in PDF

Download PDF