AWS EC2 Cost Optimization

Transparent and Fully Automated
Reduce cloud compute infrastructure cost by up to 50%

The FittedCloud Compute Optimization solution supports multiple options. Based on their specific needs, organizations can pick and choose the ideal combination of these options to optimize their compute resources.

Instance Rightsizing

Often cloud compute resources are over-provisioned by a factor or 2 or 3. FittedCloud’s Compute Optimization capability assists you in identifying AWS EC2 compute resources that are underutilized ensuring that you only provision and pay for the compute resources actually used. FittedCloud provides you with full control over the extent and timing of the optimization of your resources. FittedCloud’s Compute Optimization capability will also assist you in scaling up your resources when your applications need additional compute power or memory. All of the actions are performed seamlessly and in a secure way without any performance degradation or application disruption.

For example, you may have provisioned an ‘m4.4xlarge’ instance (16 cores, 64GB of RAM) and your application may only use part of the CPU or memory provisioned. Our software will monitor your application usage and depending on the utilization will automatically switch it to an m4.2xlarge at a cost savings of 50% per instance or even to an m4.xlarge at a cost savings of 75%.


User-defined EC2 Scheduling/Optimization

This feature allows customers to specify simple schedules that should be used to start/stop EC2 instances. Simple scheduling works well for small environments where the EC2 instance up time is known to the customers. Scheduling options include weekdays, weekends, any day of the week. There will be no machine learning in this scheme. Regardless of the state of the instance, it will be started/stopped at the specified times.

Machine Learned EC2 Scheduling/Optimization

In deployments with a large number of instances, it is often difficult to know which instances are actually used at what time. In this mode, machine learning algorithms are used to analyze historical idle times and optimum start/stop times for EC2 instances are recommended to users. Users may accept the recommendation with a few clicks. Machine-learned policies are continuously monitored to make sure that recommended policies are accurate. If system activities are detected within the suggested start/stop times, schedules are automatically adjusted. Policies are adjusted accordingly for the future optimizations.


Machine Learned Instance switching

Machine learning algorithms are also used to analyze EC2 instances utilization pattern across time and recommend right instance type. Users may accept the recommendation to switch instance types with a few clicks. Machine-learned policies are continuously monitored to ensure recommended policies are accurate.

User-defined Instance Switching

This feature allows customers to specify simple schedules that can be used to switch instance types. The feature is helpful for users that want to reduce cost by using different types of instances at different times based on known workloads. For example, one might desire to use a high-performance instance from 9 a.m. to 11 a.m. and the low-performance instance rest of the day.


Key Features:

Automated right sizing of compute resources

Application agnostic

Support for most AWS compute instance types

Support for multiple operating systems

Real-time reports on utilization for management, compliance, cost savings

Machine learning driven automated instance switching

EC2 Optimization Solution Brief