01

The modernisation ladder

FPT Software's stance: modernise, not only lift and shift, in steps the business can absorb.

  1. 01Rehost
    • Move as is with replication tools
    • The fastest way out of a data centre
    • Right-size after the move
    Same application, cloud infrastructure
  2. 02Replatform
    • Managed databases, caches, queues
    • Containers on managed Kubernetes
    • A basic CI/CD pipeline
    Less to operate
  3. 03Re-architect
    • Monolith to macro, mini, then micro services
    • Fully automated DevSecOps pipeline
    • Event-driven and serverless for spiky load
    Independent services, faster releases
  4. 04Cloud-native
    • Provider-managed services by default
    • Operations handed to the platform
    Optimised cost and speed of change
02

The container platform NGT builds

From commit to cluster: scanned images, GitOps deploys, approval before production, one view of every service.

Container Platform with GitOps
pushwebhookscan on pushdesired stateauto-syncmanual syncpull imagesGit · source of truthAWS CloudBuild, check, publishShared servicesDeploy · GitOpsVPC · private clusterObserveApp teamsApp repositorycode and DockerfileEnvironment repoHelm values per envAWS CodePipelinebuild and testQuality gateSonarQubeImage scanTrivy, before pushAmazon ECRAmazon Inspectorscan on pushAWS Secrets ManagerArgo CDpulls the desired stateApprovalstaging and productionAmazon EKSmanaged control planeDevelopmentauto-syncProductionmanual syncCPU node poolautoscaledGPU node poolfor ML workPrometheusmetricsLokilogsGrafanaone viewAmazon CloudWatchcluster insights

Point at a component to light its links; select it for what it does here and the practice that makes it right on AWS.

Point at a component to see what it does; select it for the best practice.Open the architecture
03

Database playbooks

Databases decide most timelines. These playbooks come from FPT Software's delivered migrations.

  • Oracle to PostgreSQLora2pg or AWS SCT for schema and PL/SQL, DMS for data; licences tied to hardware decide the host first.
  • SQL Server and SybaseAWS SCT and DMS, Azure DMS, or SSMA for Sybase with Qlik and Data Factory for batch jobs.
  • Into Cloud SQLMySQL, PostgreSQL and SQL Server through Database Migration Service, with continuous replication.
  • SAP and Oracle rehostedProvisioned IOPS tuned for the workload; every integration point mapped before cutover.
Database Migration to Managed Services
full load + changesby nameat cutoverOn premisesSource databasesAWS CloudMigration · until cutoverVPC · private subnetsAvailability Zone 1Availability Zone 2Cutover and runPostgreSQL / MySQLSQL ServerOracleApps that use themAWS Direct ConnectDMS SchemaConversionchange of engine onlyAWS DMSAmazon Aurora / RDSprimaryMulti-AZ standbyRead replicaAmazon RDS for SQLServerprimaryMulti-AZ standbyAmazon Route 53AWS Secrets ManagerAWS BackupAmazon CloudWatch

Point at a component to light its links; select it for what it does here and the practice that makes it right on AWS.

Point at a component to see what it does; select it for the best practice.Open the architecture
04

A typical replatform: Java to managed Kubernetes

One of the most common moves in enterprise estates, step by step.

1 of 6

Read the code for cloud blockers

Find what will break in a container before anyone changes a line.

  • Hard-coded paths, local file writes, in-process schedulers, session state
  • FleziLift reads the repository against 21 written cloud-readiness rules and cites file and line
SourcesNGT Service Offering v2.0 (Mar 2025)NGT Service Offering v3.2 (Jul 2026)NGT DevOps–AI Service Offering (Oct 2026)NGT Azure Service Offering v1.0 (Dec 2024)NGT Google Cloud Service Offering v0.1NGT Cloud Operation Service Offering v0.2 (Jan 2025)FPT Software AWS Cloud Journey Service Offering V3FPT Software Cloud Transformation Capabilities (2024)FleziLift method and runbook (Oct 2026)