Spot instances are unused capacity in a public cloud infrastructure that can be utilized at a deep discount when available. And for companies like Chegg, a Spot by NetApp customer, they’re the key to substantial cost savings. How substantial? Chegg, an online educational services provider, managed to cut Amazon Elastic Compute Cloud (EC2) costs by 70%.
The concept of Spot instances is simple. Cloud providers like Amazon Web Services (AWS), Microsoft Azure, and Google Cloud reserve large chunks of cloud storage and bandwidth to meet the needs of their on-demand customers. When there’s more supply than demand, the excess stands idle and turns to waste. Public cloud providers cut down on this waste by offering these spot instances (also called Spot virtual machines, or VMs) at a discount—often as much as 90% below on-demand pricing.
But there’s a catch. When an on-demand customer requires that excess capacity, these attractively priced Spot instances can disappear with as little as two minutes’ notice. Therefore, managing the use of Spot instances effectively requires automation to maintain reliability without sacrificing operational efficiency.
How Chegg uses Elastigroup for spot instance optimization
To solve the problem of disappearing spots, Chegg uses Spot by NetApp’s Elastigroup to automate spot instance provisioning and management. Elastigroup proactively scales clusters of virtual machines to meet capacity, scaling, and availability needs, using a cost-optimized balance of resource types and purchasing options. In addition to Elastigroup, Spot by NetApp offers a suite of cloud cost optimization tools designed to help partners optimize cloud infrastructure and reduce costs.
Spot Elastigroup monitors applications running on Spot instances and recognizes when a Spot is about to be terminated. Before termination, Elastigroup automatically shifts the application to a similar Spot instance, avoiding downtime without IT admin intervention.
For Chegg, a typical business with finite financial and human resources, Spot helped free up cash flow and engineering person-hours. Steve Evans, VP of Engineering Services at Chegg, maintains that “Chegg’s successful adoption of microservices and containers, in large part, can be attributed to Spot by NetApp keeping our infra cost and management to a bare minimum.”
The key to Spot’s automated performance is sophisticated machine learning and analytics that predict interruptions and automatically replace instances. With artificial intelligence firmly in control, partners can rely on Spot’s solutions to manage workloads, avoid costly overprovisioning, and reduce cloud infrastructure costs.
With the help of these automated services, Chegg’s operations team was able to free itself from managing infrastructure configurations. Chegg deployed Elastigroup with step-by-step guidance from their Spot support and customer success managers and soon realized a quick time-to-value and ongoing savings.