Speed, cost-efficiency, and agility are hallmarks of migration to the cloud
Posted by Ken Corless, on February 16, 2017
The business case for moving a company’s IT applications, including data analytics, to the cloud is compelling. It’s relatively fast and unquestionably cost-efficient when compared to building or expanding an in-house analytics infrastructure. It also offers a company agility in terms of developing hypotheses about how and where to use analytics in a specific organization, and then executing the resulting analytics strategy. Can your data be secure in the cloud? That’s the big question for many CIOs—and perhaps the biggest barrier to making the move.
The answer may be clearer today than it has been in the past. The top cloud service providers have built facilities, infrastructure, and architectures that employ some of the most sophisticated physical and virtual security measures in the world. Their business models cede data ownership and control to the customer—your company. They provide ample tools and guidance to help you secure and manage the data. At the end of the day, it’s your data and you control access.
The cloud providers have gone to great lengths to give you the tools to protect your data as you see fit. These tools give you the ability to protect your data at least as well as you currently do in your own data center. A key concept to understand is this: Moving your data to the cloud does not make it inherently secure; you must use the tools and processes established by the cloud provider and your existing infrastructure to protect your data.
The move to cloud can also be a catalyst to improve the security posture across the enterprise. A US financial institution recently completed its cloud strategy and came to the conclusion that the company should be encrypting all sensitive data in the cloud, whether in transit or at rest. During that discussion, the company realized that in today’s world, encryption everywhere is a practice that should be implemented in any data center.
Here are several considerations for migrating your analytics program to the cloud:
- Look fear in the face. It’s important to understand widespread concerns about securing data in the cloud, but not to be paralyzed by them.
- Reality check. Cloud service providers today offer the ability to create an environment that may be more secure than your own data centers.
- By the numbers. The longstanding best practices in data protection that you may already use in your own data centers are effective both on-premise and in the cloud (e.g. principle of least privilege, defense on failure).
- The devil’s in the details. There are important differences between all of the major cloud service providers in their shared security models. Make sure your teams understand the steps that need to be taken by each party in order to secure your data.
- Contractual terms are different. Today’s cloud providers have different service models and therefore different commercial terms than legacy hosting providers. IT teams must educate legal, risk, and procurement groups on the differences and nuances of what services are provided and how they are provided in the cloud, as well as the governance structure in place to assure quality and flexibility.
- Data regulation and compliance. Cloud providers are investing heavily to deal with their customers’ data regulation and compliance needs, and companies have discovered that compliance with cross-industry regulations, such as PCI or HIPAA, is possible in the cloud. Cloud providers are even beginning to bend to the needs of certain verticals (e.g., physical site inspections for financial regulatory bodies). Data sovereignty (where the data physically resides) is also increasingly in control of the cloud customer.
- The sky’s the limit. Once you’re comfortable with the data security issue, you can turn your attention to the many advantages of moving your analytics program to the cloud.
- Cloud-native isn’t just for websites and mobile apps. Uber, Netflix, and Snapchat get a lot of attention for using cloud platform services for building better, faster, cheaper apps. However, all three of the major cloud providers have invested heavily in building out platform services around big data and machine learning capabilities, thereby improving time to market dramatically for analytics-centric solutions. Companies like Coca-Cola1 and Redfin2 are creating business opportunities with cloud-native analytics.
In terms of developing analytics hypotheses, the cloud offers a cost-effective, iterative, and fail-fast environment that can work both quickly and effectively. Data security will always be a concern, of course. But using available tools from cloud service providers and effective practices long established in the IT world can help address these concerns.
Is your company thinking about moving your analytics capabilities to the cloud? Have you already started? I’d like to hear from you.