Logistics Firm — Cloud Migration
A three-month, zero-data-loss migration of a 40-vehicle Gwalior logistics company's on-prem server room to AWS — cutting infrastructure cost 38%, lifting shipment-tracking uptime to 99.9% and establishing a cross-region disaster-recovery posture the firm never had.
The Challenge
The logistics company ran its entire shipment-tracking stack, ERP and analytics on a single on-prem server room at its Gwalior head office — four aging rack servers and a UPS that was never quite big enough. Power cuts were a weekly fact of life in the area, and even with the generator the room would drop offline often enough that the operations team kept a spreadsheet of "outage minutes" against each month's billing cycle. Uptime was hovering around 94%.
The business cost of those outages was mounting. Shipment tracking went dark during peak dispatch windows — drivers couldn't be rerouted, customers couldn't be updated, and the customer-support phone lines would light up within minutes of any downtime. There was no disaster recovery: no off-site backups, no failover, no tested runbook. The hardware itself was past warranty, with spare parts increasingly hard to source, and the power and cooling bill for the server room was a recurring line item that grew every summer.
Leadership also knew they had outgrown the architecture. The tracking portal had no auto-scaling, so end-of-month billing runs and festival-season dispatch spikes would crawl. They needed elastic capacity, real disaster recovery, and a predictable cost curve — but they couldn't afford to lose a single shipment record or take the operation offline for more than a maintenance window.
"Our server room was a liability. Every power cut meant angry customers, drivers stuck without route updates, and a manual reconciliation exercise the next morning. We knew the cloud was the answer — we just couldn't risk the move." — K____, Operations Head
Our Solution
Curio opened with a two-week infrastructure audit — inventorying every server, application, database and integration in the server room, mapping dependencies, profiling peak load, and classifying each workload by criticality and migration risk. The audit produced a phased migration plan that prioritized the shipment-tracking system first (the revenue-critical path), then the ERP, then analytics and reporting — never moving two coupled systems in the same window.
We designed the target architecture on AWS ap-south-1 (Mumbai) for low latency to the Gwalior office and the fleet's mobile tracking devices, with EC2 instances behind auto-scaling groups to handle dispatch peaks, RDS for PostgreSQL with multi-AZ replication for the tracking database, and S3 for document and shipment-image storage. A Site-to-Site VPN connected the warehouse to the VPC so the on-prem barcode scanners continued to work unchanged. CloudWatch was configured for end-to-end monitoring, alerting the operations team on shipment-queue depth, API latency and instance health — not just CPU.
Disaster recovery was built in from day one. AWS Backup handles automated daily snapshots with cross-region replication to a second AWS region, giving the firm a tested 4-hour recovery-time objective (RTO) and 15-minute recovery-point objective (RPO) — a posture they had never had on-prem. The migration itself used a cutover model: each system ran in parallel for a defined window, was validated against the on-prem source of truth, and only then was the legacy component decommissioned. Decommissioned hardware was disposed of with certified data-wipe certificates.
-
Infrastructure AuditTwo-week inventory of servers, apps, databases and integrations; dependency mapping; workload criticality classification.
-
Cloud Architecture DesignAWS ap-south-1 target design — EC2 auto-scaling, RDS multi-AZ, S3, Site-to-Site VPN, CloudWatch, AWS Backup + cross-region DR.
-
Phased MigrationTrack system → ERP → analytics, parallel-run cutover with validation against on-prem source of truth at each step.
-
Decommission & MonitorCertified data-wipe and hardware disposal, CloudWatch alerting on shipment-queue depth, API latency and instance health.
-
DR Setup & TrainingAWS Backup + cross-region replication, DR runbook, failover drill with the operations team and go-live sign-off.
Results & Impact
The migration landed on its three-month timeline with zero shipment records lost and no unplanned downtime during business hours. Infrastructure cost — accounting for the retired server-room power, cooling, hardware refresh reserve and the on-call burden the IT team had been carrying — dropped 38% on a run-rate basis. Tracking uptime climbed from roughly 94% to 99.9%, and the customer-support team reported a sharp fall in "where is my shipment?" calls during the first month after go-live.
The disaster-recovery posture is the change the leadership team values most. A tested 4-hour RTO and 15-minute RPO, with cross-region replication, means a server-room fire, a flood or a ransomware attempt is no longer an existential event. Auto-scaling absorbs festival-season dispatch spikes without anyone having to provision ahead, and CloudWatch alerts mean the operations team knows about a shipment-queue backup before the customers do.
"Curio moved us off a server room that was a constant source of worry onto AWS without losing a single shipment record or missing a dispatch window. Tracking uptime is now 99.9%, our costs are down, and for the first time we actually have a disaster-recovery plan we've tested. The team knew logistics, not just cloud." — K____, Operations Head
Services Used
Cloud Deployment
AWS ap-south-1 architecture — EC2 auto-scaling, RDS multi-AZ, S3, Site-to-Site VPN and cross-region DR.
Explore serviceServer Management
CloudWatch monitoring, alerting on shipment-queue depth, certified hardware decommission and ongoing managed support.
Explore serviceIT Consultation
Infrastructure audit, phased migration strategy, DR runbook and failover-drill advisory for the operations team.
Explore service