Building a Data Platform for Hybrid Clouds

By Jay Kidd, CTO, NetApp


If you work in IT in 2013, you work for an IT service provider. It may not feel that way to the majority of IT professionals yet, but it does now to CIOs. The past few years of “which cloud” debates – public, private, Amazon, Azure, SaaS – have settled into the Hybrid Cloud model. This aligns with a CIO’s view of an application portfolio to serve their business and the industry view of building a range of technologies delivering IT in compelling new ways. 


CIOs care about delivering applications and managing information that their organizations depend on to operate. They will pragmatically choose the delivery model for each application that balances service level, cost, and control. Some applications will be purchased in the Software-as-a-Service model. Some will be run by cloud service providers that offer virtual private clouds or public clouds on familiar platforms. Some will be done in the hyperscalar cloud services at Amazon, Microsoft, or Google. And some will continue to be run in an enterprise’s own datacenter and by the CIO’s own staff.   


The unifying model CIOs now seek is to view all of these options through a service lens. What is the monthly cost? Am I getting the operational flexibility and responsiveness that I need? Am I paying for more than I need? Can I change quickly? This comes naturally to the Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS), and IT-as-a-Service (ITaaS) service providers. However, it is a new way of life for traditional internal IT staff as they race to become internal service providers and deliver private clouds with the same service levels as public clouds. The CIO’s view is that this portfolio of service options comprises their Hybrid Cloud. Every organization may not need or use all of these service options, but the vast majority will use at least two or more. 


So the challenge then moves from building clouds to managing the interaction between the clouds that comprise a CIO’s application landscape. How does information move between private, public, and hyperscale clouds? Can I get any operational consistency across these models to homogenize the services I am offering to my internal customers? Can I do disaster recovery (DR) between cloud options? What mix optimizes cost?  


This is where common technologies across clouds deliver the most value. Managing virtual machines (VMs) in the same way in a private cloud and a virtual private cloud closes the seams for operations. And having a common set of storage services enables more graceful deployment of data and provisioning of applications. Data ONTAP provides this common platform. 


Data ONTAP is the most deployed enterprise storage operating system in the world. More data is managed under ONTAP than in any other single storage architecture. Hundreds of thousands of customers use ONTAP in their enterprises. Over 175 NetApp Service Provider Partners also have deployed Data ONTAP as the storage foundation for over 300 service offerings. Data ONTAP can also be deployed in a manner to allow direct use by Amazon EC2 compute instances. It can also be run as a Virtual Storage Appliance running in a VM on an application host. Data ONTAP gets around.


These Data ONTAP systems can be interconnected to build a data platform for the hybrid cloud. Data can easily be replicated between an enterprise data center and a service provider to enable DR services. And in the DR scenario, the same storage efficiency services (do you really want your VMs to expand 300% when you move them to the cloud?), database cloning, snapshots and application integration are available which makes the operational model the same on premise or in the cloud. Backup services are already available from several service providers built on Data ONTAP. And a common data platform can enable services like automatic translation between virtual machine formats invoked when data moves. The possibilities for innovation are tremendous. 


Hybrid clouds are new, and a lot of experience is yet to be gained before the CIO’s goal of managing an integrated portfolio of IT services can be fully realized. But those customers who are building their private clouds on Data ONTAP and those service providers delivering ITaaS on Data ONTAP already have a head start on a key technology that provides a data management foundation of a Hybrid Cloud relationship.


Hear more about NetApp and cloud in the videos below: