Scale-up Spotinst Unveils New Serverless Engine for Containers and Kubernetes

Kelvion New Plate Heat Exchanger

spotinstAt KubeCon in Seattle, scale-up Spotinst has launched a new ‘serverless engine’ product called Ocean, which operates as a layer between containers and cloud virtual machines. The new serverless solution utilizes Spotinst’s Elastigroup technology.

Founded in 2015, Spotinst currently employs over 120 people and has raised $52 million in series A and B funding. Spotinst is based in San Francisco, with its main engineering branch located in Tel Aviv.

Spotinst Ocean lets developers deploy containers without having to provision, manage or scale the underlying infrastructure. The engine automatically adjusts the infrastructure based on container needs, significantly reducing costs by maximizing server utilization and leveraging excess capacity.

Ocean would ensure that containers clusters, such as K8S are running on the best possible mix of IaaS pricing model including Spot, Reserved, and On-Demand instances, yielding 70-85% less on infrastructure costs when compared to other managed solutions, freeing DevOps from management, and helping developers to deploy containers faster than ever before.

Patent-Pending Technology

The new serverless solution utilizes Spotinst’s Elastigroup technology, used by thousands of organizations worldwide already, to make sure that all current and future containers in the cluster have the capacity to run.

“Managing the infrastructure that powers containers is becoming quite complex,” said Amiram Shachar, co-founder & CEO of Spotinst. “Serverless technology is not only about FaaS. We’re excited to provide a service which allows containers to run without having to manage clusters or servers. It’s a true serverless experience.”

One of the main technological components of Ocean is ‘pseudo container reservations,’ a patent-pending discovery in the container space. It anticipates infrastructure requirements of future containerized workloads before the capacity is reached powered by machine learning. With it, applications would always have the capacity to scale “quickly and efficiently.”