Data Integration - Informatica

Informatica Perspectives

Avoiding A Cloud Data Disaster

Darren Cunningham Cloud computing is proving to be a disruptive force when it comes to how organizations procure and manage enterprise software. The web has truly democratized who can try and buy business applications and line of business (LOB) managers have never felt more empowered to make technology purchasing decisions – in many cases bypassing the IT organization altogether. But as software as a service (SaaS) application adoption grows and infrastructure as a service (IaaS) becomes viable for IT organizations, it’s apparent that many of the traditional barriers to on-premise software implementation success are no different in the cloud.

Do these issues sound familiar?

With the attraction of faster deployments, low monthly subscription pricing (CapEx vs. OpEx), and the potential to not have to get in line for limited IT resources, how are you going to ensure that your cloud strategy is a lightning rod for business improvement and not a data disaster waiting in the wings? Here are a few recommendations:

  1. Don’t Wait to Integrate. Jeff Kaplan at THINKstrategies lists "addressing growing integration complexities" as one of the Key Challenges Facing Cloud Computing in 2010 and Beyond. In many organizations, cloud applications have proliferated departmentally and a plan to integrate data between them has been an afterthought, if considered at all. Develop your application and data integration strategy up front and ensure that it is tightly aligned with your overall plans for business processes improvements and cost savings.
  2. Make Data Quality a Top Priority. Because of the “try before you buy” aspect of the SaaS model, organizations are often in a rush to roll out these easy-to-use web-based applications. But the impact of “garbage in, garbage out” is even more detrimental to success in the cloud. With no on-premise hardware to manage and maintain, the “pay for what you use” model means it’s never been easier to turn off an application if adoption is low. Just as a lack of data integration will lead to silos of data and "multiple views of the truth,” without timely, relevant, and trustworthy data, you can forget about any kind of return on your cloud and SaaS application investments.
  3. Get IT Involved Early. As cloud-applications and platforms have matured, some of the traditional IT objections (security, performance, availability, integration, customization, etc.) should now be questions that vendors should be able to answer with references and data center details without much difficulty. At the same time, IT organizations must establish new criterion for SaaS/cloud vendors as many of the historical RFP questions about technology standards and the like are less relevant when selecting the right SaaS/cloud solution.
  4. Think Big, Start Small. Develop a vision for cloud computing, but avoid a “big bang” approach. Do an audit of your in-house systems and determine which areas can benefit the most from SaaS adoption, which core systems will continue to run on premise, and which groups have already embraced cloud-based applications. Seek to establish standards and purchasing guidelines to avoid SaaS sprawl and ensure that departmental priorities, budgets, and skill sets are well understood.
  5. Beware of Point Solutions. There is an exciting wave of innovation happening around cloud computing and new companies are bringing fantastic technologies to the market every day. But in this economy, where start-up funding can be particularly difficult, be sure to do your homework when it comes to the long-term viability of your SaaS/cloud vendor and look for solutions that will grow with you as you move more and more computing to the cloud.

Whether you’re considering software, platform, or infrastructure as a service, embracing cloud computing can have a dramatic impact on your business in terms of cost-savings, resource allocation, agility and alignment. But in order to experience these benefits, companies of all sizes must have an overall cloud data integration strategy in place.

Useful Resources:

2 Comments, Comment or Ping

Reply to “Avoiding A Cloud Data Disaster”