From Hours of Downtime Risk to 20-Minute Recovery — Advanced Disaster Recovery on AWS

Mallouppas Group — one of Cyprus's largest retail and distribution groups, operating 50+ stores across the island for brands including MANGO, STRADIVARIUS, CALZEDONIA, and INTIMISSIMI — relies on an extensive on-premises IT infrastructure to run its Point-of-Sale systems, Active Directory, inventory, and supply chain applications. Any outage directly translates to lost sales across dozens of retail locations and disrupted distribution operations.

The company's traditional approach to disaster recovery was built on a duplicate physical DR site — complex, expensive, and rigid. Maintaining redundant hardware across multiple locations required significant capital expenditure for equipment, power, cooling, and dedicated IT personnel. DR drills and infrastructure maintenance regularly caused downtime, impacting store operations and logistics schedules.

The growing threat of ransomware added urgency. Mallouppas Group needed a recovery solution capable of restoring operations from a precise point in time — not just the last nightly backup. Meanwhile, scaling DR capacity for seasonal retail peaks or business expansion was slow and expensive, limiting the group's ability to grow with confidence.

LCM Go Cloud designed and implemented a comprehensive pilot-light Disaster Recovery solution on AWS, leveraging AWS Elastic Disaster Recovery Service (DRS) to continuously replicate 11 on-premises servers (20 disks, 4,797 GB) to the AWS cloud. The replication runs through 2 small EC2 instances with compressed, encrypted data transfer — keeping costs minimal while maintaining near-real-time data protection with an average daily change rate of just 2%.

The architecture includes Site-to-Site VPN connections between both the main and secondary customer sites and AWS, a segmented VPC with public and private subnets for isolated replication and recovery environments, and continuous Active Directory replication to AWS for seamless user authentication during failover. In a disaster scenario, recovery servers and replicated services can be brought online within minutes using automated recovery workflows.

The solution delivers a Recovery Time Objective (RTO) of under 20 minutes and a Recovery Point Objective (RPO) of under 10 minutes — with 7-day point-in-time snapshots for robust ransomware recovery. DR testing and maintenance happen with zero downtime to retail operations. A TCO analysis confirmed over 50% cost savings compared to any on-premises or colocation DR alternative, eliminating capital expenditure for duplicate hardware entirely.