Why any cloud migration strategy should start with the data warehouse

Published on the 03/05/2018 | Written by Peter O’Connor

Many Australasian organisations want to shift their IT infrastructure to the cloud, says Peter O’Connor, Snowflake Computing’s APAC VP of Sales. But where to begin?…

Interestingly, the best place to start is with the enterprise data warehouse (EDW). In many cases, this is the fastest and most risk-free path, with additional potential to increase revenue and position the business for future growth. Indeed, with data volumes growing at exponential rates, it’s not a matter of if you use the cloud to manage your enterprise data, but when.

Segmenting your IT infrastructure
Before setting out on a cloud migration strategy, it is worth taking some time to segment the IT infrastructure currently supporting the business. There are five categories to consider and their associated level of migration risk:

  1. Customer-facing applications: These applications are at the heart of any business. If something goes wrong, business is lost, revenues rapidly decline and people potentially lose their jobs. Risk = HIGH
  2. Internal applications: These include everything from email and CRM systems to payroll and general ledger. Each staff member relies on at least one of these services. Therefore, careful analysis must be done on integration points to ensure disruptions don’t occur during a migration to the cloud. Risk = HIGH
  3. Desktop/laptop OS and applications: There are a wide variety of ways in which these applications can be migrated to a cloud platform. Getting it right is vital to ensure day-to-day operations continue as normal. Risk = HIGH
  4. Operations monitoring and alerting: If the business has a network operation centre (NOC) in place, it will be integrated with virtually every application in use. Moving the NOC to the cloud could be a large and complex step. Risk = HIGH
  5. Reporting and analytics:  While these applications play an important role within a business, having a minor period of disruption is unlikely to constitute a disaster. Risk = LOW

Start with your data
Beginning a cloud migration project with the organisation’s EDW makes sense for a number of reasons, including:

  • No disruption to the business: Many EDW implementations are not mission critical, as opposed to enterprise applications. Traditional EDWs contain aggregations and are used for doing trend analysis over a period of time to make strategic, rather than tactical decisions. By starting your cloud journey with your EDW, you reduce risk by going to a more flexible architecture while giving the IT team early exposure to working with cloud services.
  • No disruption for internal users: When moving to the cloud, it’s important to be able to show incremental success and not add unnecessary risk. One option is to keep the existing EDW running in parallel with the new cloud EDW, giving you a built-in, fall-back plan for the early stages.
  • Lower start-up costs: The start-up costs associated with a cloud-based EDW are a fraction of those associated with an on-premise deployment. Many companies invested millions of dollars years ago on a data warehouse appliance that is now outdated. If they re-invest another huge sum of money, this will delay them getting to the cloud by another four or five years, putting them behind their competition. The cloud offers a pay-as-you-go model, allowing you to pay for what you use when you use it.
  • Rapid data growth: Data is growing at an exponential rate and many organisations are finding their EDW is unable to cope.  If you’re considering changing infrastructure platforms, it’s best to select tools that were built for today’s modern data challenges instead of legacy-based architectures. Moving to the cloud also gives the opportunity to consolidate operations and streamline business processes.
  • Enable new capabilities: There are some new analytic paradigms happening in the cloud such as machine learning and AI. Cloud-based platforms allow a business to work with both detailed and aggregated data at scales never imaged. It’s possible to run a complex analytics job with a 256-node Massively Parallel Processing (MPP) cluster for an hour, and then shut it down. Such flexibility delivers real business value.

As with any infrastructure move, the business benefits have to be clear enough that any reluctance to change can be overcome. The beauty of cloud adoption is that it’s easy to start small and scale without risking a huge investment up front.

Every business needs to see some proof before committing time and resources to move anything to the cloud and the EDW is a perfect candidate. Consider how you could begin your move to the cloud using this strategy.


Peter O’Connor is the Vice President Sales APJ at Snowflake Computing, based in Sydney, Australia.

Post a comment or question...

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Other Articles by Peter O’Connor

Delivering a data analytics program that delivers

opinion-article |December 16, 2019 | Peter O’Connor

Do you have the data warehouse to support a data driven future?…

Follow iStart to keep up to date with the latest news and views...