Informatica’s CIO Tony Young recently noted that when it comes to adopting cloud-based (SaaS) applications, “you must have a governance and architecture model to ensure appropriate decisions are made.” As Jeff Kaplan from THINKstrategies noted in an article focused on what he calls ”cloud control”, “uncontrolled use of cloud services may actually cost more than using better-planned resources.” So how do IT organizations get involved in the inevitable shift to cloud computing without being perceived as a gate-keeper or even worse, a roadblock? Here are a few suggestions:
- Develop a cloud-first strategy. If the US Federal Government can do it ,so can your organization. But like adapting any new technology, first determine if the cloud model makes the most sense for your business—chances are it does because the cloud offers tangible business and IT benefits , like low deployment costs and self-service, which are simply too compelling to ignore. In this economy, ensuring that your IT organization is able to deliver value to the business is the name of the game. It’s also about identifying and sticking to your core competencies. Does your core competency include managing an on-premise Learning Management System (LMS) or is there a better way? As one application manager I spoke with recently put it, “we’re either going to get on the cloud train or get run over by it.” All aboard!
- Establish the role of Data Architect. More and more IT organizations are establishing a dedicated data architect role to help guide their overall data management strategy. Aligned with the enterprise data management team that is responsible for data warehousing and reporting solutions, the role of the data architect is to drive the broader data strategy, without having to respond directly to the day-to-day delivery needs of the business. The data architect looks across systems and from this creates a unified vision that the data management team can execute. Establishing standard, data governance and working closely with the integration competency center (ICC), if this is in place, are key responsibilities of the data architect.
- Set up clear guidelines and standards when it comes to selecting the right cloud solution.Business groups do not always know or have the expertise to understand what should be included in a contract from a technology perspective. Data access privileges and data retention terminology is often vague or undefined. One way that IT can show business value is by getting involved in negotiating cloud contracts with vendors, especially since the benefits of cloud computing (self-service, operational versus capital expense, elastic computing, etc.) have bottom line impact. A best practice is for IT to develop a brief questionnaire for cloud vendors which lays out what to expect from a technology perspective. Work with the legal team to draft a standard set of questions before entering into an agreement. Doing so can help protect corporate assets while also avoiding potential downstream issues that should have been considered in the initial terms. You may be quite surprised at how well this collaborative approach is received by company executives.
- Don’t start from scratch.IT has expertise in negotiating. IT has expertise in connectivity and data integration. IT knows what must be there for an application or platform to be enterprise-ready. So don’t start from scratch. Talk to your peers, find checklists. When it comes to cloud applications, look at your existing application management best practices:
- What do you expect in terms of uptime/ availability?
- When do you need to get to your data?
- Does your data need to be backed up, encrypted, etc.?
- What are your account management roles/permissions?
- Don’t wait to integrate.For many IT departments, cloud security has shifted from an objection to a series of questions:
- Who can access your data in the cloud? Can developers, employees, system administrators access your data?
- Where can your data be accessed from and what are the options around this?
What continues to amaze me, however, is how the important topic of data integration is still often pushed out to some future stage in the overall project or plan. You want clouds, not silos, right? As Dave Linthicum points out, “many clouds are becoming just another set of silos that the enterprises must deal with.” Be sure to develop a cloud strategy that recognizes the importance of data integration and data quality up front. A failure to do this will result in disconnected systems, data fragmentation and multiple versions of the truth. Cloud data integration has emerged as both the on-ramp and the interstate for cloud applications and platforms. There are a range of approaches and solutions available in the market today, so the only question that remains is; what are you waiting for?
Darren Cunningham is VP of marketing for Informatica Cloud.