Skip to main content
AWS Lambda is a fully managed, serverless compute service that scales automatically with your applications. To maintain high performance and reliability, AWS enforces several resource limits. Understanding these constraints helps you design efficient, cost-effective solutions.

​
Key Service Limits

LimitMaximumNotes
Execution timeout15 minutesIdeal for short-lived tasks. Long-running batch jobs may require alternative services.
Memory allocation128 MB – 10 GBConfigurable per function. Affects CPU power proportionally.
Ephemeral storage (/tmp)Up to 10 GBTemporary read/write space that persists only during the invocation.
Concurrent executions1,000 (soft limit)Burst up to 3,000 in some Regions. Request a quota increase via the AWS Service Quotas console.
Memory allocation scales CPU and network throughput. Allocating more memory can improve performance for CPU-intensive workloads.
Ephemeral storage is not persistent. Data in /tmp is lost after the function completes.
To raise your concurrency limit permanently, open a quota increase request in the AWS Service Quotas console. Approval depends on your use case.
For a full overview of AWS Lambda service quotas, refer to the AWS Lambda Limits and Quotas.
The image lists limitations for a service, including a 15-minute runtime, 10 GB RAM and storage limits, and 1000 concurrent executions. It also provides a link for more information on limits and restrictions.
With Lambda’s primary limits on execution time, memory, storage, and concurrency covered, the next step is to implement robust monitoring and observability.

​
Next Steps: Lambda Monitoring

Stay tuned for our guide on setting up AWS CloudWatch metrics, logs, and alarms to track function performance, error rates, and invocation patterns.

Watch Video