Last week I attended the AWS event in Las Vegas and the buzz around Amazon reminded me of the excitement and interest VMware events generated 10 years ago. Even though I’ve been in this industry a long time I never get tired of seeing the continued technological advancement in IT.
There was, as you would expect, a lot of discussion around hybrid-clouds. AWS has seen strong growth from its initial customer base of startups and point deployments. Now, they clearly have set their sights on bigger fish; serving more of the mainstream enterprise market.
There is one big problem, however; the hybrid-cloud as it exists today is a myth.
Thus far, the tools and technologies to facilitate hybrid-clouds have been focused on one thing - migration. When viewed from a vendor perspective this is quite logical. The primary goal of any cloud operator is to acquire new workloads; facilitating bi-directional flow of information and applications is not in their best interest.
In some ways, Hybrid clouds mirror today’s wireless carriers. Each carrier provides tools and incentives to IMPORT customers but they offer nothing for EXPORTING data or workloads. Additionally, no tools exist to dynamically choose a wireless provider on a case-by-case basis.
Today, if a business wants to move applications to the public cloud they will find many tools to help. However, if the same business wants to create a flexible operational construct that provides the ability to dynamically operate across public and private environments much more work is needed.
One of the key requirements for a business to create a “genuine” hybrid-cloud is the deployment of a common data model and data access layer across both private and public cloud environments. This is necessary to provide data consistency in both form and content and to enable businesses to build a consistent and effective data governance model. Additionally, the data access layer must provide full virtualization across storage resources in both clouds not to mention to effectively manage data protection, performance, and data placement.
Hybrid-Cloud is a very innovative potential concept. But hybrid-cloud will not make it to prime time (and thus to the mainstream enterprise) without tools and technology that let a business seamlessly operate and manage applications across multiple environments. The lack of a foundational converged data access layer is an example of a key gap that must be filled.
What we have today are multiple clouds with varying degrees of compatibility and a number of migration tools to move applications and data from one cloud to another. The next step is to build the technology that enables seamless operations across clouds.