AWS Graviton Powers Up to 49% of Amazon EC2 Compute During Prime Day 2026
Key point
Amazon DynamoDB processed over 59 trillion requests and peaked at 192 million requests per second during the four-day event.
Details
Amazon Prime Day 2026, held exclusively for Prime members from June 23–26, 2026, relied on AWS infrastructure to handle millions of deals across more than 35 categories. A key highlight was the significant adoption of AWS Graviton, which powered up to 49% of the Amazon EC2 compute used by Amazon.com during the event.
Core Compute and Storage Metrics
The infrastructure faced massive throughput demands, with Amazon EBS peaking at over 24.8 trillion I/O operations and moving more than an exabyte of data daily. AWS Lambda handled over 2.3 trillion invocations per day, while Amazon ECS on AWS Fargate launched an average of 158.3 million tasks per day, a 47.7% increase from the previous year.
Database and Network Performance
Amazon DynamoDB processed over 59 trillion requests over the four days, maintaining single-digit millisecond responses and peaking at 192 million requests per second. Amazon Aurora processed hundreds of billions of transactions, storing 5,491 terabytes and transferring 1,194 terabytes of data. Meanwhile, Amazon CloudFront delivered over 2.1 trillion HTTP requests, a 5% increase compared to Prime Day 2025.
Messaging, Security, and Resilience
Messaging services saw extreme volume, with Amazon SNS delivering 5 trillion messages in a single day and Amazon SQS receiving a peak of 213 million messages per second. Amazon ElastiCache served over 2.3 quadrillion daily requests. On the security and operations front, AWS CloudTrail processed 3.6 trillion events (a 44% increase), and Amazon GuardDuty monitored an average of 14.08 trillion log events per hour (a 59% increase). To ensure availability, AWS ran over 44,000 AWS FIS experiments, more than six times the number conducted in 2025.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.