At VMWorld, VMware CEO Pat Gelsinger delivered the keynote address on the opening day of the conference. Along with several C-level clients, one of the main themes Gelsinger explored was how important the hybrid cloud model was to each customer, and how VMware’s new cross-cloud architecture announcement shifted the company’s strategy to enable hybrid. VMware also announced an expansion of its partnership with IBM, which would be the first global partner to host the new VMware cloud foundation.
During the keynote, VMware shared a couple of demos. First, they showed how NSX would be able to secure traffic within an AWS environment with the new VMware cross-cloud SaaS management tool. Then they showed a workload that was running in AWS in the same graphical user interface (GUI). Finally, Gelsinger said VMware can migrate that workload to another cloud, live, or to an on-prem target. But, he then noted it would take several hours to move the data – so they stopped the demo, rather than ask everyone to wait for that data movement to complete.
He just stopped – in the middle of a live demo. To wrap things up, Michael Dell joined Gelsinger on stage to talk about the importance of the hybrid, cross-cloud vision and why the full muscle of Dell technologies is behind this vision.
Hold on a minute! I sat there in the audience wondering, “Why does it take several hours to migrate the data?” Well, it’s because in order to migrate data from on location to another, such as from on-prem to cloud, you must copy all of your data first to storage in your target location. The same holds true if you are going to migrate from one cloud to another, or from one on-prem location to another. When you dig into cross-cloud architecture, you see very clearly in the documentation, blogs, release notes, best practices and reference architecture docs that you (the customer) are directly responsible for the following:
- Procuring the network bandwidth
- Paying to store at least two copies of your data (in fact, it’s more likely 6 or more with VSAN)
- Continuously moving that data in order to enable data mobility
This is not quite the flexible, hybrid cloud infrastructure that was pitched during those first few minutes of the keynote address. However, this exact challenge has been on our minds for years.
True data mobility means accessing all of your data when and where you want it, without the cost or complexity of replicating it. One of the exciting use cases that comes up regularly in our customer meetings involves attaching compute on demand to a storage service environment, in order to augment existing on-prem environments. In that scenario, the customer pays for the compute cycles when they’re needed without having to transport and store a secondary copy of the data just for insurance purposes. This is critical, as on-demand environments can be used for a variety of purposes: from universities looking at augmenting their environment during course registration before the spring and fall semesters, to insurance companies doing the same with the open enrollment period for benefits, to organizations looking to eliminate their DR facility.
Replication was a 20th century storage solution, and it’s become a 21st century storage problem. It’s also no longer necessary to unlock data mobility. Hybrid cloud infrastructure is achievable and accessible today.
Learn more about how a global storage network enables hybrid cloud.