1. Home
  2. Projects
  3. AWS Migration, File Services and Disaster Recovery

AWS Migration, File Services and Disaster Recovery

Landing zone, MGN migration waves, FSx for Windows and tested recovery

  • Lead AWS Engineer
  • AWS
  • Control Tower
  • Organizations
  • MGN
  • EC2
  • FSx
  • DataSync
  • Robocopy
On this page

Overview

A financial-services organization needed an AWS landing zone, server migration, Windows file services and a tested disaster-recovery capability. The engagement extended from discovery and foundation build through migration waves, data transfer, permission remediation and recovery drills.

Challenge

A third-party managed firewall controlled both internet and internal traffic, so the AWS design had to fit an existing vendor-operated network model. File migration also changed course after the initial transfer method failed, and migrated permissions needed remediation before users could rely on the new file service.

My role

I led discovery, authored the landing-zone and disaster-recovery designs, deployed the AWS foundation, ran MGN migration waves, designed and built FSx for Windows, coordinated the file-data migration, corrected permissions and led customer failover and failback sessions.

Architecture & approach

Simplified view of the engagement scope. Confidential details omitted.

Key components: AWS Control Tower, AWS Organizations, AWS Application Migration Service, Amazon EC2, Amazon FSx for Windows, AWS DataSync, Robocopy, Active Directory, AWS Elastic Disaster Recovery, Terraform.

What I implemented

  • Deployed the AWS landing zone and integrated it with the managed firewall architecture.
  • Troubleshot domain-controller connectivity and later reconfigured east-west traffic paths.
  • Executed MGN test and cutover waves, including off-hours migration bridges.
  • Built Amazon FSx for Windows and replaced the initial file system after a throughput issue.
  • Changed the data-transfer method from DataSync to Robocopy from an EC2 host when the original approach did not work.
  • Remediated file permissions and produced the disaster-recovery design and operating documentation.
  • Led failover and failback disaster-recovery drills with the customer.

Key decisions

  • Changed the file-migration method based on observed failure rather than continuing with an unsuitable tool.
  • Treated the managed firewall vendor as part of the operating model and validated internet, domain and east-west flows explicitly.
  • Paired recovery design with customer-run exercises so the runbook reflected actual behavior.

Security & reliability

The implementation used a governed AWS foundation, controlled network paths, managed Windows file services and documented recovery procedures. Migration and DR activities were tested with customer participants and updated based on the results.

Technologies

  • Control Tower
  • Organizations
  • MGN
  • EC2
  • FSx
  • DataSync
  • Robocopy
  • Active Directory
  • DRS
  • Terraform

Outcome

The organization received a working AWS foundation, migrated workloads, production file services with corrected permissions, and a tested failover and failback process.

Lessons & takeaways

Key takeaway

The file-services work reinforced the need to validate throughput, transfer behavior and permissions early. A successful copy is not a complete migration until access and operational performance are confirmed.