AWS DynamoDB Cost Optimization

Amazon DynamoDB is a fully-managed NoSQL database. When you create a DynamoDB table, you provision the desired amount of request capacity taking into account the amount of read and write traffic and the average size of each item. DynamoDB doesn’t provide built-in capabilities to auto-tune throughput based on application load. Customers are charged on an hourly basis for provisioned read and write throughput. Organizations have to set provisioned read and write capacity for every table and global secondary index. Setting something too low will degrade performance and increase the risk of application crashing whereas setting something too high will result in paying for unused resources. Currently, without FittedCloud, the capacity can be changed manually but requires the customer to constantly monitor and take action, which are both time consuming activities.

FittedCloud DynamoDB Dynamic Resource Optimization solution solves this problem. The solution uses machine learning algorithms to learn the usage of DynamoDB tables from past usage pattern. It then couples that intelligence with automated actions to optimize DynamoDB provisioning. The result is fully automated, optimum, DynamoDB capacity provisioning driven by machine learning algorithms.

DynamoDB Optimization can be performed in two different modes.

  1. Scheduled optimization based on user specified time windows.
  2. Machine learning algorithm driven.

FittedCloud makes DynamoDB provisioning dynamic, automated and transparent, offering potentially significant cost savings for customers.

Key Features:

Both machine learning or user specified schedule based optimization

Capacity adjustments based on CloudWatch alerts of excessive capacity utilization

Complete lights out operations

Dynamic configuration

Visualization of provisioned vs consumed capacity per table

Dashboard and reports for easy visualization, cost savings reports

DynamoDB Optimization Solution Brief