3/27/2025, 12:00:00 AM ~ 3/28/2025, 12:00:00 AM (UTC)

Recent Announcements

Amazon GameLift Servers expands instance support with next-generation EC2 instance families

Amazon GameLift Servers now supports Amazon EC2 5th through 8th generation instances, offering enhanced price-performance, efficiency, and flexibility for game server hosting. This update allows developers to leverage the latest advancements in EC2 compute, memory, and networking across three main instance families:\n

General Purpose (M-series): Balanced CPU, memory, and networking for a wide range of game workloads.

Compute Optimized (C-series): High-performance compute instances with a 2:1 memory ratio, ideal for CPU-intensive game servers.

Memory Optimized (R-Series): Optimized for high-memory workloads with an 8:1 memory ratio, supporting complex simulations and large player sessions.

Each new EC2 generation brings significant improvements:

5th Gen: Proven reliability with Intel processors with balanced performance

6th Gen: Includes AWS Graviton2 ARM-based options alongside Intel and AMD variants offering enhanced price-performance efficiency.

7th Gen: The latest evolution featuring DDR5 memory, enhanced networking, and offering significant performance gains over previous generations.

8th Gen: Cutting-edge AWS Graviton4 and Intel Xeon-based instances for demanding workloads

Customers can also choose variants with local storage (d), enhanced networking (n), and different processor architectures (Intel, AMD, Graviton – i/a/g). This update empowers developers with greater flexibility, scalability, and cost efficiency to optimize game server performance. Customers can now seamlessly transition workloads to newer EC2 generations, leveraging AWS’s continuous innovation for building, scaling, and operating multiplayer games globally. These next-generation instances are available in Amazon GameLift Servers supported regions, except AWS China. For more information on launching fleets with next-generation EC2 instances, visit the Amazon GameLift Servers documentation and EC2 Instance Types overview.

AWS Network Manager and AWS Cloud WAN now support AWS PrivateLink and IPv6

AWS Network Manager and AWS Cloud WAN now support AWS PrivateLink and IPv6 based connectivity to the management endpoint of these services. Using PrivateLink, customers can now access AWS Network Manager or AWS Cloud WAN privately on the AWS network, without going through the public Internet. Additionally, customers can now access these services over IPv6 using dual-stack endpoints.\n With AWS Cloud WAN, you can use a central dashboard and network policies to create a global network that spans multiple locations and networks, allowing you to configure and manage different networks using the same technology. The Cloud WAN central dashboard, powered by AWS Network Manager, generates a complete view of the network to help you monitor network health, security, and performance. AWS Network Manager reduces the operational complexity of managing global networks across AWS and on-premises locations. Previously, you could access AWS Cloud WAN and AWS Network Manager using public IPv4 endpoints only. With this launch, you can now access these services’ APIs/CLI privately, without going through the public Internet. Additionally, these services now support IPv6 endpoints. To learn more about AWS Network Manager, refer documentation, and for AWS Cloud WAN, refer documentation.

AWS Network Firewall adds pass action rule alerts and JA4 filtering

Today, AWS announces new features for AWS Network Firewall: The ability to generate alerts on traffic that matches pass action rules and JA4 fingerprinting support in firewall rules. AWS Network Firewall is a stateful, managed, network firewall and intrusion detection and prevention service for your virtual private cloud (VPC). These new capabilities enhance the security and visibility of your network traffic, allowing for more granular control and improved threat detection.\n The ability to generate alert log events on traffic that matches pass action rules provides enhanced visibility into your network traffic without a need to add an alert action rule before the pass action rule. This can help you detect anomalies or potential security issues in traffic that would otherwise be permitted without additional scrutiny. JA4 filtering rules enables AWS Network Firewall to analyze network traffic based on JA4 fingerprints, which are used to identify client and server applications. This feature allows for more precise traffic identification and control, helping you to better secure your network against potential threats. Pass action rule alert and JA4 filtering rules are available in all AWS Regions where AWS Network Firewall is offered. To see which regions AWS Network Firewall is available in, visit the AWS Region Table. To learn more about these new features and how to implement them in your AWS Network Firewall setup, visit the AWS Network Firewall documentation. You can start using these new capabilities today to enhance your network security posture and gain deeper insights into your VPC traffic patterns.

Amazon Bedrock Knowledge Bases now supports Amazon Opensearch Managed Cluster for vector storage

We are announcing the support of Amazon OpenSearch Managed cluster as a vector store in Amazon Bedrock Knowledge Bases. Amazon Bedrock Knowledge Bases securely connects foundation models (FMs) to internal company data sources for Retrieval Augmented Generation (RAG), to deliver more relevant and accurate responses.\n Amazon Bedrock Knowledge Bases’ native integration with vector databases allows you to mitigate the need to build custom data source integrations. With this launch, you can use OpenSearch managed cluster as the vector database to take advantage of the suite of features available in Bedrock Knowledge Bases. This integration adds to the list of vector databases supported by Bedrock Knowledge Bases, including Amazon OpenSearch Serverless, Amazon Aurora, Amazon Neptune Analytics, Pinecone, MongoDB Atlas, and Redis. The OpenSearch Managed cluster integration for Amazon Bedrock Knowledge Bases is now generally available in all existing Amazon Bedrock Knowledge Base and Opensearch service regions. To learn more, refer to the Knowledge Bases documentation.

AWS Marketplace introduces new seller experiences for Machine Learning products

Today, we are excited to announce a set of improvements to the seller management experience for Machine Learning (ML) products in AWS Marketplace. Sellers can now quickly publish and update ML listings with a new self-service experience in the AWS Marketplace Management Portal. Additionally, sellers can now create, view, and manage private offers for ML products through a guided step-by-step process in the AWS Marketplace Management Portal, making it easier for them to extend custom pricing terms to their customers. To help improve operational efficiency, sellers can now utilize AWS Marketplace Catalog APIs to automate creating and updating ML product listings and private offers.\n These three features help streamline the process of listing and managing ML products in AWS Marketplace, offering a more seamless experience for ML sellers and enabling them to bring their innovative ML solutions to customers faster and more efficiently. With the new self-service experience, sellers can perform all listing creation and management actions and publish private offers in minutes, all without requiring help from AWS Marketplace support. With access to APIs, ML sellers can automate their listing creation and update processes by integrating directly with AWS Marketplace from within their model publishing pipelines. To get started, visit the Machine Learning product page in the AWS Marketplace Management Portal. To learn more, access the AWS Marketplace Seller Guide.

Amazon EKS now enforces upgrade insights checks as part of cluster upgrades

Today, Amazon Elastic Kubernetes Service (EKS) announced a new control to prevent accidental cluster upgrades when issues are already detected that may impact application compatibility with the next Kubernetes version. This feature leverages EKS upgrade insights and is significant step towards giving cluster administrators confidence with Kubernetes version upgrades.\n EKS upgrade insights automatically scan clusters against a list of potential Kubernetes version upgrade impacting issues such as deprecated Kubernetes API usage. EKS periodically updates the list of insight checks to perform, based on evaluations of changes in the Kubernetes project, as well as changes introduced in the EKS service along with new versions. With this new control, EKS will prevent you from upgrading the EKS clusters if there are any Kubernetes version upgrade impacting issues surfaced by EKS upgrade insights. Once the upgrade impacting issues are resolved, you will be able to upgrade the Kubernetes version of your cluster. EKS has also introduced an override flag which you can use to bypass upgrade insights checks on upgrades, which can useful for example in dev environments. This feature is available in all AWS Regions, except the AWS GovCloud (US) Regions. To learn more visit the EKS documentation.

Amazon EC2 P5en instances are now available in US East (N. Virginia) and Asia Pacific (Jakarta)

Starting today, Amazon Elastic Compute Cloud (Amazon EC2) P5en instances powered by NVIDIA H200 GPUs are available in US East (N. Virginia) and Asia Pacific (Jakarta) regions. These instances are optimized for generative AI and high performance computing (HPC) applications.\n P5en instances feature 8 H200 GPUs which have 1.7x GPU memory size and 1.4x GPU memory bandwidth than H100 GPUs featured in P5 instances. P5en instances pair the H200 GPUs with high performance custom 4th Generation Intel Xeon Scalable processors, enabling Gen5 PCIe between CPU and GPU which provides up to 4x the bandwidth between CPU and GPU and boosts AI training and inference performance. P5en, with up to 3200 Gbps of third generation of EFA using Nitro v5, shows up to 35% improvement in latency compared to P5 that uses the previous generation of EFA and Nitro. This helps improve collective communications performance for distributed training workloads such as deep learning, generative AI, real-time data processing, and high-performance computing (HPC) applications. To address customer needs for large scale at low latency, P5en instances are deployed in Amazon EC2 UltraClusters, and provide market-leading scale-out capabilities for distributed training and tightly coupled HPC workloads. With these additional regions, P5en instances are now available in the US East (N. Virginia, Ohio), US West (Oregon), Europe (Spain) and Asia Pacific (Jakarta, Mumbai, Seoul, Tokyo and Seoul) AWS Regions and US East (Atlanta) Local Zone us-east-1-atl-2a in the p5en.48xlarge size. To learn more about P5en instances, see Amazon EC2 P5en Instances.

Announcing 3 new features on Connected Mobility Solution on AWS

Today AWS announced role-based access control, multi-account and multi-region support, and Fleet Management Portal (Preview) for the Connected Mobility Solution (CMS) on AWS. Role-based access control enables customers to manage and restrict access to developer portal of CMS based on security roles assigned to users. In addition, customers can use CMS to deploy AWS and partner provided software components in multiple accounts and multiple regions based on customer’s organization structure and various lifecycle stage of the features. Lastly, customers can use the CMS Fleet Management Portal to visualize fleet data collected using AWS IoT FleetWise, making it easier to assemble a holistic view of fleet events, and easily integrate and visualize AWS and AWS partner provided software components.\n You can restrict access to CMS’s developer portal by defining roles and permissions with appropriate level of access within your organization, ensuring access is granted to users based on the ‘principle of least privilege’ to reduce security risks. You can easily deploy AWS and AWS partner provided connected mobility software components across multiple accounts and regions to enable alignment with your organizational requirements and enhance scalability and resilience. You can accelerate the build of your fleet management solution, visualize consolidated fleet data, and quickly integrate insights and analytics solutions provided by AWS and connected mobility partners using the CMS-provided API Gateway and Smithy Framework.

CMS is available in the following AWS regions: Asia-Pacific (Tokyo, Sydney), Europe (Frankfurt, Ireland), US-East (N.Virginia, Ohio), US-West (Oregon).

Visit CMS on AWS Product Page to learn more.

Amazon DynamoDB now supports percentile statistics for request latency

Amazon DynamoDB now supports percentile statistics for the SuccessfulRequestLatency Amazon CloudWatch metric. The percentile statistic enables you to understand the latency distribution of your successful requests to DynamoDB, complementing the existing average, minimum, and maximum statistics.\n The SuccessfulRequestLatency metric only measures latency which is internal to the Amazon DynamoDB service - client side activity and network trip times are not included. It’s normal to see some variability in this metric. When analyzing your latency, it’s best to consider your end-to-end latency which includes client side activity. To factor in your client side activity, you can enable latency metric logging in your AWS SDK. The new percentile statistic is available in all commercial AWS Regions, the AWS GovCloud (US) Regions, and the China Regions. To learn more about the percentile statistic and troubleshooting latency on DynamoDB, see the following:

DynamoDB Metrics and Dimensions in the DynamoDB Developer Guide

Troubleshooting latency issues in Amazon DynamoDB in the DynamoDB Developer Guide

AWS CloudFormation now supports targeted resource scans in the IaC generator

Today, AWS CloudFormation introduced a new resource scanning workflow for the CloudFormation IaC generator, further simplifying the process of generating Infrastructure-as-Code (IaC) templates for existing resources in your AWS account. IaC generator allows you to onboard existing resources to CloudFormation in three easy steps. First, you initiate a scan of resources in your AWS account. Second, you select resources for template generation and review suggestions for related resources. Third, a CloudFormation template is generated for selected resources. You can then import resources into a CloudFormation stack, download the template for deployment, or convert the template into a CDK app in your preferred programming language, such as TypeScript or Python.\n With this launch, you can specify the resource types that IaC generator will cover in the resource scanning step. Instead of scanning all resources by default, you can now focus only on the resources relevant to your workload, reducing scan time and effort. This improves the efficiency of the template generation process and streamlines iterative workflows, such as migration of a prototype workload to CloudFormation. To get started, open the AWS CloudFormation Console and select IaC generator in the navigation panel. You can also use IaC generator from the AWS CLI and AWS SDK. Learn more:

User guide

The IaC generator is available in AWS Regions where CloudFormation is available.

CloudFormation documentation for Partial Scanning

Amazon Q Business now available in AWS Asia Pacific (Sydney) Region

Starting today, Amazon Q Business is available in Asia Pacific (Sydney) AWS Region. Amazon Q Business revolutionizes the way that employees interact with organizational knowledge and enterprise systems. Q Business customers in this region can get answers from enterprise RAG knowledge bases and uploaded files (e.g. pdf’s, images) and run tabular search on small tables. Customers can also get answers from LLM knowledge and generate content using their Q Business assistant. Amazon Q Business connects seamlessly to over 40 popular enterprise systems, including Amazon Simple Storage Service (Amazon S3), Microsoft 365, and Salesforce. It ensures that users access content securely with their existing credentials using single sign-on, according to their permissions, and enterprise-level access controls.\n With this regional expansion, Amazon Q is now available in the following regions: US East (N. Virginia), US West (Oregon), Europe West (Ireland), and Asia Pacific Southeast (Sydney) AWS Regions. To learn more about the Amazon Q Business features available in this region, go to Q Business service regions. For more information, see Amazon Q Business.

Amazon VPC IP Address Manager is now available in two additional AWS Regions

Amazon Virtual Private Cloud IP Address Manager (Amazon VPC IPAM) that makes it easier for you to plan, track, and monitor IP addresses for your AWS workloads, is now available in Asia Pacific (Thailand) and Mexico (Central) Regions.\n Amazon VPC IPAM allows you to easily organize your IP addresses based on your routing and security needs, and set simple business rules to govern IP address assignments. Using VPC IPAM, you can automate IP address assignment to Amazon VPCs and VPC Subnets, eliminating the need to use spreadsheet-based or homegrown IP address planning applications, which can be hard to maintain and time-consuming. With this expansion, Amazon VPC IPAM is available in all AWS Regions, including China (Beijing, operated by Sinnet), and China (Ningxia, operated by NWCD), and the AWS GovCloud (US) Regions. To learn more about IPAM, view the IPAM documentation. For details on pricing, refer to the IPAM tab on the Amazon VPC Pricing Page.

AWS Blogs

AWS Japan Blog (Japanese)

AWS News Blog

AWS Compute Blog

AWS Developer Tools Blog

AWS DevOps & Developer Productivity Blog

Integration & Automation

AWS Machine Learning Blog

Open Source Project

AWS CLI

AWS CDK