ATW Resilience: Optimizing Live Chat Performance with SecureKloud's Cloud Factory Model

Published date : Jan 2024

Executive Summary

SecureKloud's Cloud Factory Model, featuring Automatic Target Weights (ATW) in Application Load Balancer, effectively resolved connectivity challenges for Optimy's Live Chat solution. ATW ensures continuous monitoring, automatically addressing issues and enhancing overall user experience.

About the client

The client is a leading Canadian company offering a cutting-edge sales-focused Live Chat solution that helps organizations to enhance their buying experience via one-way or two-way video chat, collaborative browsing, etc. With the AI-powered Live Chat solution, the users can deliver a customized and personalized chat experience for their customers. The client is the tech division of a 15-year industry leader, delivering face-to-face retail sales programs.


Global Locations


Certified Cloud Architects


Years of Cloud Experience


Cloud Transformations

Business Challenge

Optimy team reported that their customers were facing some failure on the application connectivity. Some of the time application goes on unresponsive status.

SK team tried to investigate the reported issue on infrastructure perspective while some instances handle traffic smoothly, others experience performance issues and errors. During that time issue led to a degraded user experience and even lost sales. Even though servers have increased, customer faced the error occasionally.

Our Solution

Based on the investigation, some instances handle traffic smoothly, others occasionally experience performance issues and errors. This led to a degraded user experience and even lost the live chat connectivity from the application.

  • We have recommended the option called Automatic Target Weights (ATW) in Application Load Balancer. This feature automatically adjusts the weight of individual targets (instances) behind the ALB based on their health metrics.
  • ATW's anomaly detection analyzes the HTTP return status codes and TCP/TLS errors to identify targets with uneven ratio of errors compared to other targets in the same target group.
  • When ATW identifies anomalous targets, it reduces traffic to the under-performing targets and gives a larger portion of the traffic to targets that are not exhibiting these errors. When the partial failure decreases or stops, ALB will slowly increase traffic back onto these targets.
  • For the first level, we have implemented this option in the QA environment for customer validation.

Business Outcome

After enabling the ATW with anomaly detection option now it monitors and maintains application availability by automatically handling temporary issues with individual instances in the target group.