This website uses cookies. By continuing to browse, you agree to our Privacy Policy.

How Zubale Increased Dev Efficiency and Productivity Using Komodor

Company Size:

101 - 250 employees

Industry:

e-commerce

Komodor Installation:

40 nodes, 3 clusters, 1748 services

About Zubale

Zubale enables brands and retailers to win in eCommerce. They partner with a wide range of retailers from fashion brands, supermarkets, specialty stores, and pharmacies to integrate Zubale’s technology with theirs and complete the end-to-end fulfillment of their eCommerce orders. With Zubale, retailers can compete and win against the B2C App aggregators while keeping their most important asset: Their customers.

The Challenge

Zubale was fast to adopt Kubernetes when it hit the scene, following the promise of a scalable, self-healing, and cost-efficient infrastructure. However, the Zubale team quickly learned that it takes a lot of time and resources to unlock the full potential of K8s, and even more work to maintain a reliable and stable system. 

They had a clear goal in mind: to build a DevOps culture and automate everything. To achieve this, they set out to build a scalable, automated, and transparent platform on GKE. One of their primary objectives was to enable development teams to take ownership of their services while still following the principle of least privilege access. By doing so, they could promote a sense of accountability and responsibility amongst the development teams. 

Additionally, they sought to reduce the MTTR from complex issues related to failed deployments, config changes, and stateful sets in a K8s environment. By streamlining and automating their processes, they could minimize the risk of downtime and improve the overall efficiency of their operations.

The Problem

Prior to its transformation, Zubale had several critical issues with its development process. Firstly, they lacked a development environment, leading to the pushing of changes directly to production or staging environments. This not only introduced unnecessary risk but also made it difficult to test and iterate on features. Furthermore, the senior developers were in charge of infrastructure, leading to a bottleneck in the deployment process. This made it challenging for the development teams to take ownership of their services and work efficiently. Finally, there was no role-based access control (RBAC) in place, which meant that developers either had full access or no access at all. This not only created security vulnerabilities but also hindered the ability to collaborate and work together effectively.

The Solution

With the help of Komodor, Zubale were able to reduce unnecessary escalations to SRE and DevOps, resulting in saving a significant amount of time every week.  Komodor’s solution offered context to failure by providing events and enriched data, enabling developers to have a deeper understanding of the inner workings of K8s and resolve many minor incidents independently without involving the organization’s K8s experts.

As not all the developers are well-versed in Kubernetes, interacting with their K8s apps via Komodor’s Platform proved to flatten the learning curve in more meaningful discussions with the DevOps/SRE team – instead of asking for a general ‘fix’ to an issue, they come with specific requests and can essentially all speak the same language.

Furthermore, Komodor’s platform also provided robust RBAC and OOTB policies to prevent any damage or unaudited changes to the system, giving the infra team control and visibility over everything happening in their clusters – giving devs access and freedom to roam, while maintaining governance and enforcing best practices.

This enabled the teams to view their system’s health in a single pane of glass, thereby reducing the need for extensive investigation and troubleshooting, and helping infrastructure teams to quickly pinpoint common issues during deployments and daily operations.

Overall, Komodor’s Platform helped Zubale’s engineering teams to improve their productivity and efficiency while also reducing downtime.

Download this Case Study in PDF

Download PDF