Amazon EC2 Capacity Manager is a new centralized solution that simplifies the monitoring, analyzing, and managing of capacity usage across all AWS accounts and regions from a single interface. It aggregates capacity data, provides insights into usage patterns, optimizes costs by identifying underutilized resources, and eliminates the need for custom automation or manual data collection. The service also offers detailed dashboards and export capabilities for enhanced analysis and decision-making.

Amazon EC2 Auto Scaling now allows users to force cancel instance refreshes immediately, enhancing control during critical updates by enabling rapid deployment changes without waiting for ongoing instance activities. This feature is particularly useful in emergencies when service disruptions occur. Users can implement this by setting the WaitForTransitioningInstances parameter to false in the CancelInstanceRefresh API call.

Amazon Web Services (AWS) has announced a price reduction of up to 45% for its NVIDIA GPU-accelerated Amazon EC2 instances, including P4 and P5 instance types. This reduction applies to both On-Demand and Savings Plan pricing across various regions, aimed at making advanced GPU computing more accessible to customers. Additionally, AWS is introducing new EC2 P6-B200 instances for large-scale AI workloads.

AWS has announced updates to the pricing and usage model for Amazon EC2 instances powered by NVIDIA GPUs, including the introduction of savings plans for P6-B200 instances and significant price reductions for P5, P5en, P4d, and P4de instances. These changes, effective June 2025, aim to enhance accessibility to advanced GPU computing across various global regions.

Amazon EC2 now allows users to automatically delete underlying EBS snapshots when deregistering AMIs, streamlining the process and reducing storage costs. This feature is available to all customers at no additional cost across all AWS commercial regions. Users can deregister AMIs through the EC2 Console, CLI, API, or SDK.

Pinterest faced challenges with EC2 network throttling that affected application reliability and performance, particularly during user sequence serving and EC2 instance migrations. By enhancing network performance monitoring through AWS Elastic Network Adapter metrics and implementing techniques like fine-grained S3 rate limiting, Pinterest improved visibility and mitigated network issues, ensuring better performance for its online services.

Amazon EC2 has launched M4 and M4 Pro Mac instances, designed to enhance performance for Apple platform development by utilizing Apple M4 and M4 Pro chips. These instances offer improved build performance, increased memory capacity, and seamless integration with AWS services, making them ideal for continuous integration and delivery pipelines. Both instance types support macOS Sonoma and provide low-latency storage for better caching and performance.

Amazon has announced the general availability of EC2 C8gn instances powered by Graviton4 processors, which offer up to 600Gbps network bandwidth, the highest among EC2 network optimized instances. These instances are designed for demanding network-intensive workloads and provide enhanced compute performance compared to previous Graviton3-based instances. C8gn instances are now available in select AWS regions and can be launched via various AWS tools.

Amazon EC2 Capacity Manager is now generally available, allowing customers to monitor and manage their EC2 capacity across multiple accounts and regions from a single interface. The service provides dashboards for usage insights, historical trends, and optimization opportunities, all without additional costs. It is also integrated with existing systems through data export capabilities.