Underutilized Write Capacity on a DynamoDB Table

Service Category

Databases

Cloud Provider

AWS

Service Name

Amazon DynamoDB

Inefficiency Type

Underutilization

Explanation

Provisioned capacity mode is appropriate for workloads with consistent or predictable throughput. However, when write capacity is significantly over-provisioned relative to actual usage, it results in wasted spend. This inefficiency is especially common in dev/test environments, legacy systems, or workloads that have tapered off over time but were never adjusted.

Relevant Billing Model

In Provisioned Capacity Mode, DynamoDB charges hourly for:

Write Capacity Units (WCUs) and Read Capacity Units (RCUs) based on the provisioned level, not actual consumption
Additional services such as backups, streams, and global tables also generate usage-based charges

Write and read capacity is billed continuously, regardless of whether it is fully used. Over-provisioning WCUs or RCUs leads to avoidable cost.

Detection

Identify DynamoDB tables using Provisioned capacity mode
Review utilization history over a 14-day or longer period to assess average write throughput
Determine whether actual write usage is consistently below the provisioned capacity
Check for the absence of throttling or bursty usage patterns that would require excess headroom
Evaluate whether Auto Scaling is enabled and if current minimum thresholds are still appropriate
Confirm the table's environment (e.g., production vs. test) and business criticality
Engage with workload owners to validate whether capacity can be safely reduced or if On-Demand is more appropriate

Remediation

Reduce provisioned write capacity to better match observed usage, either manually or by tuning Auto Scaling settings. For workloads with highly variable or low average usage, consider switching to On-Demand mode, which offers pay-per-request pricing and removes the need to manage throughput provisioning.

Relevant Documentation