Optimize latency-sensitive workloads with Amazon EC2 detailed NVMe statistics
Compute Blog
This article explains how to use Amazon EC2 detailed NVMe statistics to optimize latency-sensitive workloads with locally attached instance storage.
- EC2 Nitro instances now offer 11 comprehensive per-second granularity metrics for NVMe instance storage
- Metrics include queue length, IOPS, throughput, and IO latency histograms with IO size-specific distributions
- Access statistics via nvme-cli command or CloudWatch agent for real-time visibility
- Scenario 1: Use "Performance Exceeded" metric to identify when workloads exceed instance storage limits
- Scenario 2: Use IO-size specific histograms to diagnose how large writes affect read performance
- Provides detailed troubleshooting examples and sample monitoring scripts
EC2 detailed NVMe statistics enable precise performance monitoring and bottleneck identification for high-performance applications requiring ultra-low latency storage.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Sep 16
2025
2025
Amazon EC2 supports detailed performance stats on all NVMe local volumes
Nov 3
2025
2025
Amazon CloudWatch Agent adds support for NVMe Local Volume Performance Statistics
Nov 25
2025
2025
Performance benefits of new Amazon EC2 R8a memory-optimized instances
Dec 2
2025
2025
Announcing Amazon EC2 Memory optimized X8i instances (Preview)
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.