AI BriefingThe Pursuit of 99.999%: Surviving with Multi & Hybrid Clusters
Key point
Kakaopay Securities built a multi and hybrid cluster environment using AWS, IDC, and KakaoCloud to achieve 99.999% service availability.
Details
In the securities industry, 99.999% service availability is not just a goal but a prerequisite for maintaining customer trust. To solve the risk of a single-cloud-level failure causing a complete service outage when operating a single cluster, Kakaopay Securities introduced a multi and hybrid cluster structure.
The service currently runs on a purpose-fit combination of three platforms: AWS EKS, IDC Kubernetes, and KakaoCloud Kubernetes Engine (KC KE). This reduces cloud dependency and secures infrastructure flexibility.
The core strategy is as follows:
- Deployment layer: Services are deployed in parallel across multiple clusters through the wallga deployment pipeline. Developers only need to extend the
platformfield in the service specification to enable multi-cluster deployment without any additional work. - Traffic control: GSLB (Global Server Load Balancing) is used to unify the load balancers of multiple clusters under a single domain. If a specific cluster fails, GSLB detects this and automatically reroutes traffic to a healthy cluster.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
KT Cloud Aims to Cut Deployment Failure Root Cause Analysis Time by 90% Using MCPA Choice for Seamless Evolution in Large-Scale Kubernetes Environments: kt cloud's Chkk-Based Cluster Management