Amazon Elastic Compute Cloud (EC2) is a powerful infrastructure-as-a-service solution that provides a broad set of instance types and offers tremendous flexibility with respect to how they are configured, used, and purchased.
Organizations often struggle to determine the optimal EC2 instance types to use for their workloads. Even though there are millions of ways to configure and purchase EC2, the decision of exactly what types of instances to buy is often made manually, imprecisely, and infrequently. This causes operational risks, poor instance utilization, and inflated bills.
By applying machine learning to this challenge, our Cloud-Learning Optimization Engine—Cloe—is able to analyze workload patterns to precisely determine which instance types should be applied to each cloud workload. Using detailed supply and demand models, and applying benchmarks to all computations, Cloe is uniquely able to generate advanced cross-family recommendations, while at the same time driving high levels of automation.
Cloe is “Reserved Instances-Aware” and considers your standard RI coverage, protecting existing investments. If your instances have existing standard RI coverage, the recommendations will be deferred until the current RI expires.See How Our Machine Learning Can Optimize Your EC2 Environment
The process of optimizing the instances being purchased can be automated, replacing a tedious and error-prone task with a fact-based, API-driven process. Densify enables apps running in EC2 to become self-aware of their resource requirements, and can leverage Terraform templates to automate the process of moving workloads to the best-matched EC2 instance types and sizes. With this next generation of automation, time is saved, operational risk is reduced, and the alignment of workloads with the optimal instance types saves an average of 42% on your EC2 cloud costs.Assess Opportunities for Optimization in Your Own EC2 Environment