Case Study

August 4, 2017

Canifa – Highway to Cloud


Case Studies



CANIFA is a fashion brand in Vietnam, founded in July 29, 2007. After 10 years of running, CANIFA is now  a top-notch player in the industry, well-known for their high quality products, especially their fall/winter collection.

The company soon realized the online shopping trend will come to this country, and have built its own ecommerce website to extend the reach to customers. During sales seasons, the website saw ever-increasing visitors, frequent down times, and numberous complaints on loadind speed. All these factors raised the question on how to keep it both highly scalable & cost effective.

1. The challenges

During consultation process, we can see various challenges that Canifa was facing:

  • Current system is built on physical infrastructure
  • Physical server cost is too expensive
  • Current system cant serve enourmous number of requests when running promotion campaigns
  • Extreme difficulty in database backup and restoring
  • No disaster recovery solutions
  • Caching system is ineffective, and unable to handle auto failover.

2. Our solution

We persuaded Canifa to go cloud, designed system architecture, performed migration & deployment, setup autoscaling system, and provided monthly performance reports.

The architecture we applied for Canifa includes:

  • Virtual Private Cloud (VPC) : to manage Private Network likes traditional VPN
  • MySQL: Using RDS MySQL – Multi AZ, managed service of AWS, IT guys don’t need to care about the setup & install problem. They only need focus to monitoring. RDS also very easy to change instance size in minutes.
  • Webservers: Using EC2s for Magento Ecommerce website
  • Disaster recovery: Using multiple zones in region singapore.
  • EC2s & ELB: Auto-scaling, highly available and scalabile system.
  • Elasticache multi AZ for Memcached Cache Engine
  • Storage: S3, using SDK for PHP language

3. The outcome

Benefits from going cloud can be seen quickly, especially during the Black Friday campaign that came right after the completion of deployment:

  • No records of system outage
  • The system auto-scales up/down 5-6 times/day
  • ELB: work perfectly, serving unlimited requests, never die endpoint.
  • Database: no worries failover problems by using RDS Multi-AZ and automatically backup snapshots daily
  • Storage: no files loss up till now.
  • No upfront


  • Reply

    Hi, this is a comment.
    To get started with moderating, editing, and deleting comments, please visit the Comments screen in the dashboard.
    Commenter avatars come from Gravatar.

  • Reply



Copyright © 2018 - OSAM. All Rights Reserved.