NetApp Expands Its New Spot Tech For Dynamic Cloud Compute Instances

‘Spot has AI and machine learning in the platform to predict further in advance when a spot instance will disappear. It understands the customer application and resources needed and will use on-demand, reserved and spot instances to lower the costs,’ says John Woodall, vice president of engineering West at NetApp partner General Datatech.

NetApp is expanding its newly acquired Spot Elastigroup technology for automatically migrating cloud-based data to the lowest-cost compute instances with the ability to work with the new Microsoft Azure Spot Virtual Machines.

NetApp in June acquired Spot, a developer of technology that drives continuous optimization of workloads running on a public cloud while ensuring that those applications follow customer service-level agreements (SLAs) and service-level objectives (SLOs).

Spot does this in part by taking advantage of compute spot instances available from Amazon Web Services, Azure and Google Cloud Platform. Until now, Spot Elastigroup has supported Azure’s low-priority virtual machines but now works with Microsoft Azure Spot Virtual Machines.

[Related: 10 Key Takeaways From NetApp CEO George Kurian: Cloud, Coronavirus And Growth]

Spot in general finds the lowest-cost compute instances within AWS, Google Cloud Platform and Azure and migrates data to those instances dynamically to help lower the cost of using cloud compute, said Amiram Shachar, vice president and general manager of Spot by NetApp.

The three cloud providers offer three types of compute instances, Shachar told CRN. Customers can use on-demand pricing ,which charges by the hour and is the most expensive; reserved pricing, which requires purchasing instances ahead of use; or spot pricing, where excess capacity is available at the lowest cost but which can be taken away without warning, he said.

“The Spot software lays across all three pricing models to provide the highest availability at the lowest price,” he said. “When customers move to the cloud, they are looking for elastic pricing and the ability to scale up or down. This is where our application shines.”

While the ability to use spot compute instances to support demand with little advance notice is nice, the caveat is the instance may disappear two or three minutes later, making them not very useful for production workloads historically, said John Woodall, vice president of engineering west at General Datatech, a Dallas-based solution provider and longtime NetApp channel partner.

“Spot has AI and machine learning in the platform to predict further in advance when a spot instance will disappear,” Woodall told CRN. “It understands the customer application and resources needed and will use on-demand, reserved and spot instances to lower the costs.”

Other tools help monitor spot instances, Woodall said. “However, NetApp Spot not only does the analysis, but will act on your behalf and provide four-nines [99.99 percent] SLAs,” he said.

The COVID-19 coronavirus pandemic has customers dealing with new challenges related to their cloud budgets, which are now typically higher than planned, Woodall said.

“NetApp Spot becomes an easy discussion to address these two customer questions: how to save money in the cloud and how to adopt more cloud if the cost decreases,” he said. “Spot provides a tremendous value proposition. It lowers the cost, provides SLAs and simplifies the cloud”

Going forward, solution providers should expect to see news related to increased NetApp Spot support for Azure containers and the Kubernetes service during the upcoming NetApp Insight conference being held virtually late next month, Shachar said.