Organizations use their Data warehouses to drive analytics and generate insights. To aid this process, data warehouses are designed to hold historic data in de-normalized structures. These de-normalized structures help simplify querying from these large structures. Netezza, Teradata, Exadata are some of the most common on-premise data warehouses. Typical problems customers face with on-premise data warehouses include scalability, high degree of management and expensive upfront compute and storage costs. These factors have contributed to customers shifting to Cloud based Data warehouse solutions.
AWS Redshift is the Cloud Data warehouse offering from AWS. This is a petabyte scale fully managed data warehouse on Cloud, which in conjunction with other services on AWS, enables organizations to gleam new insights of their data. Redshift uses the familiar ANSI SQL, enabling users to seamlessly migrate and integrate with the AWS ecosystem.
The Infosys Redshift migration strategy couples our expertise in implementing data warehouse and Infosys accelerators, to help enterprises accelerate the journey to Cloud data warehouse.
Cloud Data Warehouse – Migration Strategies:
Strategies for migration to Redshift Cloud Data Warehouse.
Re-Host
(Lift and Shift)
- Moves applications without changes and cloud optimization
- Moves quickly to meet business objectives
- Save roughly 25-30 % of costs compared to on-premises.
Re-platform
(Lift, Refine & Shift)
- Moves applications with few cloud optimization
- Moves the Databases to cloud DB which are fully managed services.
- Remediates/Optimizes the ETL code to Cloud DB.
- Remediates the reporting to point cloud DB.
- Architects cloud-based work-flow orchestration
Re-architect / Refactoring
- This strategy tends to be the most expensive, however most beneficial in long term.
- Re-architects the applications and develop using cloud-native features.
- Driven by a strong business case to add capabilities, scale, or performance which is difficult to achieve in current platform.
- Enables migration from a monolithic architecture to a service-oriented architecture to boost agility or improve business continuity
Fig 1: Data Warehouse Migration to AWS Redshift - Patterns
Components in a Data Warehouse migration:
A Data Warehouse migration consists of migrating/rebuilding several components such as Schemas (Tables, Views), Historic data Migration and Delta data append, building ETL components and rewiring consumption components to the new Data warehouse.
- Teradata
- Netezza
- Oracle Exadata
- Talend
- Informatica
- SQL Server
- Microstrategy
- Qlik
- Business Objects
- Tableau
AWS SCT / AWS Database Migration Service/ Accelerators
AWS SCT / Partners Solution / Internal Tool/ Manual
Database Jobs / ETL Migration
AWS SCT / Partners Solution / Internal Tool / Manual
Reporting / Downstream Migration
- Informatica
- Talend
- Redshift
- Business Objects
- Microstrategy
- Tableau
- Qlik
Fig 2: Components in a Data Warehouse Migration
AWS Services that help in migrating to Redshift Data Warehouse:
- AWS SCT (AWS Schema Conversion Tool) helps with converting schema from several of the commonly used data warehouses to AWS specific databases including Redshift. The converted schema can also be used with a database on an Amazon EC2 or stored as data on AWS S3.
- AWS DMS (AWS Database Migration Service) helps migrate data from source database to AWS. It supports both homogenous and heterogeneous data migration. When used with SCT, DMS can convert the schema from the source database to target and can also help migrate data from source to target.
Infosys Data Wizard:
Infosys data wizard is a comprehensive framework with a set of accelerators for seamless Data Migration.
Fig 3: Infosys Data Wizard unboxed
