As Amazon Web Services prepares to make a “significant announcement” at a press conference in San Francisco Thursday afternoon, speculation is flying about what the leading public cloud unit of Amazon.com Inc. will unveil.
As SiliconANGLE reported last week, at least part of the announcement will involve a deal with longtime rival VMware Inc., whose software allows multiple versions of the same operating software to run on the same computers. Sources close to the situation say VMware customers will be allowed to run its software both in their private data centers and on AWS’s cloud in a setup known as hybrid cloud.
But the specifics remain murky ahead of the announcement by AWS Chief Executive Andy Jassy (above), which will be live-streamed starting at 1:30 p.m. Pacific. The event could involve a number of related announcements, such as the ability to run VMware’s NSX virtual networking software.
But the key impact could be on big data workloads, one of the fastest-growing workloads in computing today. They’re starting to move to the public clouds. Wikibon, the analyst community owned by the same company as SiliconANGLE, has identified the three most likely scenarios for how an AWS-VMware deal will affect big data workloads (subscription required for the full report with details of each scenario).
First, some background: On-premise and public cloud computing have largely grown up independent of each other. Although there are some high-level similarities in developer abstractions or administrative processes, creating hybrid applications has been extremely challenging, particularly between VMware on-premises and AWS in the public cloud.
These are the key scenarios that could come out of the partnership:
Hybrid scenario No. 1: AWS offers a separate VMware cloud alongside AWS. In this scenario, a pool of VMware-managed Infrastructure as a Service resources exists within the AWS cloud. Beyond common billing, there is limited integration between VMware in the cloud and AWS.
Hybrid scenario No. 2: AWS manages vSphere virtual machine in addition to the native Xen VM. In this scenario, AWS can manage its own Xen-based VMs as well as VMware’s native vSphere. AWS would probably accomplish this by hosting vSphere VMs inside Xen VMs.
Hybrid scenario No. 3: AWS extends the No. 2 scenario and manages on-premises edge appliances running the AWS stack with vSphere support. In this scenario, AWS provides VMware customers with an appliance similar to Microsoft Corp.’s on-premises Azure Stack. But in this case, the appliance would run infrastructure that AWS could manage in a hybrid configuration.
The upshot of all the scenarios is that big data professionals are finally going to have a more meaningful coexistence or a migration path from on-premises “private clouds” to the dominant and more fully automated public cloud run by AWS.
Prioritizing what to deploy on whatever path emerges starts with reviewing the type of workload and then the hybrid cloud configuration. Ephemeral workloads should see the largest total-cost-of-ownership savings, regardless of the hybrid cloud scenario. But the deepest integration between VMware and AWS, the third scenario, should yield the greatest savings.
George Gilbert is big data and analytics analyst at Wikibon.