Moving Database Applications to the Cloud: How Hard Can It Be? (Part 1)

Cloud Express: The Journey to the Cloud Begins

10.03.18

Migrating to the Cloud: Top Considerations when Replatforming Applications

09.19.18

Faster time to market increasingly drives companies of all sizes to move their legacy data warehouse applications to the cloud. Other factors include: cost reduction, operations simplification, scalability and elasticity. However, all these benefits could be offset by the biggest challenge Information Technology (IT) departments face, namely, choosing the best path to moving legacy data warehouse applications to the cloud.

How hard can a Data Warehouse Migration be?

Migrating an existing data warehouse to the cloud is a complex process of moving schema, data, applications and Extract Transform and Load (ETL) jobs. There are many strategic questions that need to be answered as well before embarking in this journey:

  • What is the best way to move based on your current environment?
  • Which applications should be moved?
  • Which applications should stay on-premises?
  • What is the best strategy: lift and shift, replatform, or refactor?
  • Do you do all cloud-native? Public cloud? Private cloud? A mix?
  • What are the consequences of delays or failure to replatform mission-critical applications?

The complexity increases significantly if a decision is made to restructure the database schema, rebuild the data pipelines (ETL jobs) or modernize existing applications while moving legacy data warehouse applications to the cloud.

Most Common Approaches to Moving Legacy Data Warehouse Applications to the Cloud

These are the most common paths to moving legacy data warehouse applications to the cloud:

  • Lift and Shift: This is defined as moving the application without any modifications to the cloud. This approach is a faster and less resource-intensive process. The downside is that it does not benefit from cloud-native features like elasticity. Depending on the cloud provider and the setup this option could be costlier than replatforming or refactoring (see below) but more economical than running on-premise.
  • Replatforming applications to the cloud: This is a middle ground between Lift and Shift and refactoring. While a slower migration path than Lift and Shift, it is less complex than refactoring. This path allows workloads to take advantage of base cloud functionality and cost optimization, without the level of resource commitment required for refactoring.
  • Refactoring: This is the most time consuming of all three approaches. It involves rearchitecting and recoding an existing application to take advantage of cloud-native frameworks and functionality. It is also the one with the highest risk and lowest success rate.

Keep in mind that a typical enterprise data warehouse contains a large amount of data describing many business subject areas. Moving an entire data warehouse in a single pass (A Flash cut or Cold Turkey approach1) is usually not realistic. Incremental migration (Incremental step-by-step or Chicken Little approach1) is the smart approach. Migrating incrementally becomes imperative when undertaking these kind of projects.

The complexity increases significantly if a decision is made to restructure the database schema, rebuild the data pipelines (ETL jobs) or modernize existing applications while moving legacy data warehouse applications to the cloud.

Adaptive Database Virtualization Accelerates Adoption of Cloud Databases

Adaptive Database Virtualization (ADV)2 solves the problem of moving data warehouses to the cloud effectively and at a fraction of the cost of other approaches. Datometry© Hyper-Q™ implements ADV to enable applications running on legacy on-premise databases to be re-pointed to a cloud database such as: Azure SQL DW™, Amazon Redshift™, Snowflake Elastic Data Warehouse™ or on-premise using Pivotal Greenplum©.

How does it work? Hyper-Q™ intercepts the SQL queries as they are executed by the application and in real time either translates, transforms or emulates the SQL queries executed from the language used by the application.

moving your data warehouse applications to the cloud
Hyper-Q handles SQL queries syntactic differences (translation), semantic differences (transformation) and compensates for missing functionality (emulation) between the source and target database. Each one of these features will be covered in a separate blog post in this series.

Learn more about Hyper-Q technology in this downloadable white paper: Rapid Adoption of Cloud Data Warehouse Technology Using Datometry Hyper-Q.

References

(1) Migrating Legacy Systems Gateways, Interfaces & The Incremental Approach. Michael L. Brodie, Michael Stonebraker, Morgan Kaufmann Publishers, Inc.
(2) Rapid Adoption of Cloud Data Warehouse Technology Using Datometry Hyper-Q
Lyublena Antova, Derrick Bryant, Tuan Cao, Michael Duller, Mohamed A. Soliman, Florian M. Waas. ACM SIGMOD 2018

7 things every IT leader should know before moving analytics to the cloud
Download PDF

About Alex Infanzon

Alex Infanzon holds a Master of Science in New Generation Computing from The University of Exeter in the United Kingdom and a B. A. in Computer Science from The Anahuac University in Mexico. He has over 30 years of professional experience in consulting, pre-sales and management roles at Westinghouse, Informix Software, Sun Microsystems, Composite Software, Dun and Bradstreet, EMC Greenplum and most recently the SAS Institute. He has worked with fortune 500 companies in different vertical industries, both in a selling and consulting capacity. Alex has extensive experience in architecting and implementing Business Intelligence, Data Management, Enterprise Information Integration, Fuzzy Matching and High Performance Computing solutions.

Continue Reading