Publicado el — Deja un comentario

AWS HealthImaging now supports OpenID Connect (OIDC) authentication for DICOMweb APIs

AWS HealthImaging now supports OAuth 2.0-compatible identity providers for authentication of DICOMweb requests using OpenID Connect (OIDC). With OIDC authentication, you can manage secure access to DICOM resources using your organization’s standard procedures for creating, enabling, and disabling user accounts.

With this launch, you can now use existing identity providers (IdPs)—such as Amazon Cognito, Okta, or Auth0—to issue JSON Web Tokens (JWTs) that authorize secure access to your DICOMweb endpoints. This launch makes it simpler to integrate AWS HealthImaging into existing medical imaging applications and expands HealthImaging’s support of DICOMweb standard interfaces that rely on OAuth 2.0-compatible authentication. Support for OIDC is limited to DICOMweb REST API requests. HealthImaging includes native support for AWS Identity and Access Management (IAM) users and roles for authentication of all API requests.

Support for OpenID Connect (OIDC) is available in all AWS Regions where AWS HealthImaging is generally available: US East (N. Virginia), US West (Oregon), Asia Pacific (Sydney), and Europe (Ireland).

To learn more, visit Using DICOMweb with AWS HealthImaging.

 

​AWS HealthImaging now supports OAuth 2.0-compatible identity providers for authentication of DICOMweb requests using OpenID Connect (OIDC). With OIDC authentication, you can manage secure access to DICOM resources using your organization’s standard procedures for creating, enabling, and disabling user accounts. With this launch, you can now use existing identity providers (IdPs)—such as Amazon Cognito, Okta, or Auth0—to issue JSON Web Tokens (JWTs) that authorize secure access to your DICOMweb endpoints. This launch makes it simpler to integrate AWS HealthImaging into existing medical imaging applications and expands HealthImaging’s support of DICOMweb standard interfaces that rely on OAuth 2.0-compatible authentication. Support for OIDC is limited to DICOMweb REST API requests. HealthImaging includes native support for AWS Identity and Access Management (IAM) users and roles for authentication of all API requests. Support for OpenID Connect (OIDC) is available in all AWS Regions where AWS HealthImaging is generally available: US East (N. Virginia), US West (Oregon), Asia Pacific (Sydney), and Europe (Ireland). To learn more, visit Using DICOMweb with AWS HealthImaging.  

Publicado el — Deja un comentario

AWS IoT SiteWise now supports retraining of anomaly detection models

Today, AWS announced new capabilities for native anomaly detection in AWS IoT SiteWise. This release includes automated model retraining, flexible promotion modes, and exposed model metrics, all designed to enhance the anomaly detection feature.

The automated retraining capability allows models to be automatically retrained on a schedule ranging from a minimum of 30 days to a maximum of one year, eliminating the need to manually retrain models. This feature ensures that models stay up-to-date with changing equipment conditions or configurations, thereby maintaining optimal performance over time.

Additionally, flexible promotion modes give customers the choice between service-managed and customer-managed model promotion. Automatic promotion enables AWS IoT SiteWise to evaluate and promote the best-performing model without customer intervention, while manual promotion allows customers to review comprehensive, exposed model metrics—including precision, recall, and Area Under the ROC Curve (AUC)—before deciding which model version to activate. This flexibility allows choice between a hands-off or human oversight approach. 

Multivariate anomaly detection is available in US East (N. Virginia) , Europe (Ireland) , and Asia Pacific (Sydney) AWS Regions where AWS IoT SiteWise is offered. To learn more, read the launch blog and user guide. 

 

​Today, AWS announced new capabilities for native anomaly detection in AWS IoT SiteWise. This release includes automated model retraining, flexible promotion modes, and exposed model metrics, all designed to enhance the anomaly detection feature. The automated retraining capability allows models to be automatically retrained on a schedule ranging from a minimum of 30 days to a maximum of one year, eliminating the need to manually retrain models. This feature ensures that models stay up-to-date with changing equipment conditions or configurations, thereby maintaining optimal performance over time. Additionally, flexible promotion modes give customers the choice between service-managed and customer-managed model promotion. Automatic promotion enables AWS IoT SiteWise to evaluate and promote the best-performing model without customer intervention, while manual promotion allows customers to review comprehensive, exposed model metrics—including precision, recall, and Area Under the ROC Curve (AUC)—before deciding which model version to activate. This flexibility allows choice between a hands-off or human oversight approach.  Multivariate anomaly detection is available in US East (N. Virginia) , Europe (Ireland) , and Asia Pacific (Sydney) AWS Regions where AWS IoT SiteWise is offered. To learn more, read the launch blog and user guide.   

Publicado el — Deja un comentario

Amazon IVS now supports private ingest via interface VPC endpoints

Amazon Interactive Video Service (Amazon IVS) now supports media ingest via interface VPC endpoints powered by AWS PrivateLink. With this launch, you can securely broadcast RTMP(S) streams to IVS Low-Latency channels or IVS Real-Time stages without sending traffic over the public internet. You can create interface VPC endpoints to privately connect your applications to Amazon IVS from within your VPC or from on-premises environments over AWS Direct Connect. This provides private, reliable connectivity for your live video workflows.

Amazon IVS support for media ingest via interface VPC endpoints is available today in the US West (Oregon), Europe (Frankfurt), and Europe (Ireland) AWS Regions. Standard AWS PrivateLink pricing applies. See the AWS PrivateLink pricing page for details.

To learn more, please visit the Amazon IVS private ingest documentation page.

 

​Amazon Interactive Video Service (Amazon IVS) now supports media ingest via interface VPC endpoints powered by AWS PrivateLink. With this launch, you can securely broadcast RTMP(S) streams to IVS Low-Latency channels or IVS Real-Time stages without sending traffic over the public internet. You can create interface VPC endpoints to privately connect your applications to Amazon IVS from within your VPC or from on-premises environments over AWS Direct Connect. This provides private, reliable connectivity for your live video workflows.
Amazon IVS support for media ingest via interface VPC endpoints is available today in the US West (Oregon), Europe (Frankfurt), and Europe (Ireland) AWS Regions. Standard AWS PrivateLink pricing applies. See the AWS PrivateLink pricing page for details.
To learn more, please visit the Amazon IVS private ingest documentation page.  

Publicado el — Deja un comentario

BrainKey y Microsoft Azure: Avanzar en la investigación de la salud cerebral con IA responsable

septiembre 10, 2025

BrainKey y Microsoft Azure: Avanzar en la investigación de la salud cerebral con IA responsable

Doctora habla con un paciente mientras le muestra una tablet

Por: Microsoft para Startups.

A veces, las condiciones médicas y los diagnósticos más desafiantes requieren un cambio de perspectiva. Las nuevas empresas innovadoras que utilizan IA y Microsoft Azure han encontrado nuevas formas de introducir esas perspectivas en la práctica de la medicina, todos los días.

Mientras Owen Phillips, director ejecutivo de BrainKey, neurocientífico capacitado en Stanford, observaba cómo la salud cognitiva de su madre se deterioraba con rapidez, se encontró en un lugar donde ninguna cantidad de experiencia académica podría haberlo preparado para experimentar, y ninguna solución existente parecía capaz de diagnosticar su condición.

A pesar de sus profundas conexiones en el campo de la medicina, su condición permaneció sin diagnosticar, hasta que la tecnología que construía en su startup, BrainKey, al final ayudó a descubrir el verdadero problema: una afección tratable llamada hidrocefalia normotensiva.

Ese momento se convirtió en el catalizador de la misión de BrainKey: capacitar a los pacientes y médicos con información más temprana y precisa sobre la salud del cerebro por medio de IA e imágenes médicas.

Detrás de todo lo que hacemos en BrainKey está este impulso para ayudar a las personas a evitar el tipo de incertidumbre y dolor por el que pasó mi familia. Queremos que la salud del cerebro sea más accesible, comprensible y procesable.

—Owen Phillips, director ejecutivo de BrainKey

Construir una IA responsable que detecte patrones que otros pasan por alto

La «clave» de BrainKey es el uso de IA avanzada para analizar resonancias magnéticas y detectar cambios sutiles en el cerebro, a menudo antes de que aparezcan los síntomas. ¿Cómo? Por medio de modelos que se entrenan en un vasto conjunto de datos normativos recopilados de clínicas de todo el mundo. Estos conjuntos de datos masivos permiten que la tecnología BrainKey estime la «edad cerebral» e identifique signos tempranos de deterioro neurológico.

Pero BrainKey no se detiene en el análisis. Han construido una interfaz de visualización tridimensional que también permite a los pacientes explorar sus propios escáneres cerebrales en un formato intuitivo y fácil de usar. Este enfoque en la experiencia del usuario, junto con el rigor científico, es lo que distingue a BrainKey.

No solo construimos herramientas para médicos. Creamos herramientas que ayudan a las familias a comprender lo que les sucede a sus seres queridos.

—Nathan Strong, director de tecnología de BrainKey

Escalado con Azure: una plataforma creada para la precisión, el rendimiento y la privacidad

Para respaldar su creciente base de usuarios y cargas de trabajo intensivas en datos, BrainKey recurrió a Microsoft Azure. Su infraestructura usa Azure Kubernetes Service (AKS) para el proceso escalable, Azure Blob Storage para el control seguro de datos y Terraform para la administración flexible de la infraestructura.

Lidiamos con cargas de trabajo en ráfagas: algunas horas son tranquilas; otras están llenos de análisis. El escalado automático de Azure y la compatibilidad con nodos heterogéneos han sido fundamentales para administrarlo de manera eficiente.

—Nathan Strong, director de tecnología de BrainKey

La seguridad y el cumplimiento también son las principales prioridades. La arquitectura de BrainKey es compatible por completo con HIPAA, con datos cifrados en reposo y en tránsito. Para optimizar aún más la solución, BrainKey se ha integrado con Nuance PowerShare, una plataforma propiedad de Microsoft, para facilitar el intercambio de datos con los proveedores de atención médica.

Uso del ecosistema de Microsoft para respaldar la innovación en la atención médica

El recorrido de BrainKey con Microsoft comenzó a través del programa Microsoft for Startups en 2019, donde recibieron créditos, soporte técnico y acceso a una red global.

Desde entonces, la asociación se ha profundizado, con el equipo de Iniciativas Digitales y Estratégicas (DSI, por sus siglas en inglés) de Microsoft que desempeña un papel fundamental en la optimización de la arquitectura de BrainKey y la exploración de oportunidades de comercialización.

El soporte de Microsoft ha sido fenomenal. Desde revisiones técnicas hasta introducciones estratégicas, nos han ayudado a avanzar más rápido y de manera más inteligente.

— Owen Phillips, neurocientífico formado en Stanford

BrainKey ahora ha comenzado a explorar cómo integrarse con Microsoft Azure Marketplace, lo que permite a los proveedores de atención médica adquirir su solución a través de los canales existentes de Microsoft. Esto abre la puerta a una adopción más amplia y una colaboración más profunda con los equipos de cuentas de atención médica de Microsoft.

De las clínicas a Microsoft Azure: Apoyar la salud del cerebro en la nube

Con un acuerdo reciente firmado con uno de los proveedores de imágenes por resonancia magnética (MRI, por sus siglas en inglés) más grandes de los Estados Unidos, BrainKey se ha comenzado a preparar para escalar con rapidez. ¿Su objetivo? Llevar su plataforma de salud cerebral impulsada por IA a clínicas de todo el país y, de manera eventual, del mundo.

Construimos la infraestructura ahora para respaldar un crecimiento masivo. Azure nos brinda la escalabilidad que necesitamos para que esto esté disponible para todos, no solo para unos pocos. Más adelante, incluiremos Brainkey en Azure Marketplace. Este es un hito importante para nosotros, ya que hará que nuestra tecnología sea más accesible para los proveedores de atención médica de todo el mundo y permitirá una integración perfecta en los flujos de trabajo clínicos existentes.

— Owen Phillips, neurocientífico formado en Stanford

A medida que BrainKey continúa su crecimiento, su misión se mantiene clara: ayudar a millones de personas a comprender mejor sus cerebros, recibir diagnósticos más tempranos y vivir vidas más saludables. Y Microsoft ayuda a garantizar que la tecnología desempeñe un papel transformador para llevarlos allí.

Empiecen a usar Microsoft for Startups hoy mismo

The post BrainKey y Microsoft Azure: Avanzar en la investigación de la salud cerebral con IA responsable appeared first on Source LATAM.

 

​The post BrainKey y Microsoft Azure: Avanzar en la investigación de la salud cerebral con IA responsable appeared first on Source LATAM.  

Publicado el — Deja un comentario

Amazon CloudFront adds ECDSA support for signed URLs

Amazon CloudFront now supports Elliptic Curve Digital Signature Algorithm (ECDSA) for signed URLs and signed cookies, providing customers with enhanced performance and security for content access control. This addition gives customers the flexibility to choose between RSA and ECDSA cryptographic algorithms based on their specific security and performance requirements.

Previously, CloudFront only supported RSA based encryption algorithms to create signed tokens. ECDSA offers several advantages over traditional RSA signatures, including faster signature generation and verification, smaller signature sizes that result in shorter URLs, and equivalent security with smaller key sizes. This makes ECDSA signed URLs and signed cookies particularly beneficial for high-volume applications, mobile environments, and IoT devices where processing efficiency and bandwidth optimization are critical.

ECDSA support with signed URLs and signed cookies is available in all edge locations. This excludes Amazon Web Services China (Beijing) region, operated by Sinnet, and the Amazon Web Services China (Ningxia) region, operated by NWCD. There is no additional charge to utilize this feature. To learn more about restricting content delivered with Amazon CloudFront, visit the CloudFront documentation. 

 

​Amazon CloudFront now supports Elliptic Curve Digital Signature Algorithm (ECDSA) for signed URLs and signed cookies, providing customers with enhanced performance and security for content access control. This addition gives customers the flexibility to choose between RSA and ECDSA cryptographic algorithms based on their specific security and performance requirements. Previously, CloudFront only supported RSA based encryption algorithms to create signed tokens. ECDSA offers several advantages over traditional RSA signatures, including faster signature generation and verification, smaller signature sizes that result in shorter URLs, and equivalent security with smaller key sizes. This makes ECDSA signed URLs and signed cookies particularly beneficial for high-volume applications, mobile environments, and IoT devices where processing efficiency and bandwidth optimization are critical. ECDSA support with signed URLs and signed cookies is available in all edge locations. This excludes Amazon Web Services China (Beijing) region, operated by Sinnet, and the Amazon Web Services China (Ningxia) region, operated by NWCD. There is no additional charge to utilize this feature. To learn more about restricting content delivered with Amazon CloudFront, visit the CloudFront documentation.   

Publicado el — Deja un comentario

TwelveLabs’ Marengo Embed 2.7 can now be used for synchronous inference in Amazon Bedrock

Amazon Bedrock now supports synchronous inference for TwelveLabs’ Marengo 2.7, expanding the capabilities of this multimodal embedding model to deliver low-latency text and image embeddings directly within the API response. This update enables developers to build more responsive, interactive search and retrieval experiences while maintaining the same powerful video understanding capabilities that have made Marengo 2.7 a breakthrough in multimodal AI.

Since its introduction to Amazon Bedrock earlier this year, Marengo 2.7 has transformed how organizations work with video content through asynchronous inference—ideal for processing large video, audio, and image files. The model generates sophisticated multi-vector embeddings, enabling precise temporal and semantic retrieval across long-form content. Now with synchronous inference support, users can leverage these advanced embedding capabilities for text and image inputs with significantly reduced latency. This makes it perfect for applications such as instant video search where users find specific scenes using natural language queries, or interactive product discovery through image similarity search. For generating embeddings from video, audio, and large-scale image files, continue using asynchronous inference for optimal performance.

Marengo 2.7 with synchronous inference is now available in Amazon Bedrock in US East (N. Virginia), Europe (Ireland), and Asia Pacific (Seoul). To get started, visit the Amazon Bedrock console and request model access. To learn more, read the blog, product page, Amazon Bedrock pricing, and documentation. 

 

​Amazon Bedrock now supports synchronous inference for TwelveLabs’ Marengo 2.7, expanding the capabilities of this multimodal embedding model to deliver low-latency text and image embeddings directly within the API response. This update enables developers to build more responsive, interactive search and retrieval experiences while maintaining the same powerful video understanding capabilities that have made Marengo 2.7 a breakthrough in multimodal AI. Since its introduction to Amazon Bedrock earlier this year, Marengo 2.7 has transformed how organizations work with video content through asynchronous inference—ideal for processing large video, audio, and image files. The model generates sophisticated multi-vector embeddings, enabling precise temporal and semantic retrieval across long-form content. Now with synchronous inference support, users can leverage these advanced embedding capabilities for text and image inputs with significantly reduced latency. This makes it perfect for applications such as instant video search where users find specific scenes using natural language queries, or interactive product discovery through image similarity search. For generating embeddings from video, audio, and large-scale image files, continue using asynchronous inference for optimal performance. Marengo 2.7 with synchronous inference is now available in Amazon Bedrock in US East (N. Virginia), Europe (Ireland), and Asia Pacific (Seoul). To get started, visit the Amazon Bedrock console and request model access. To learn more, read the blog, product page, Amazon Bedrock pricing, and documentation.   

Publicado el — Deja un comentario

Amazon MSK Connect is now available in Asia Pacific (Malaysia)

Amazon MSK Connect is now available in the Asia Pacific (Malaysia) Region. MSK Connect enables you to run fully managed Kafka Connect clusters with Amazon Managed Streaming for Apache Kafka (Amazon MSK). With a few clicks, MSK Connect allows you to easily deploy, monitor, and scale connectors that move data in and out of Apache Kafka and Amazon MSK clusters from external systems such as databases, file systems, and search indices. MSK Connect eliminates the need to provision and maintain cluster infrastructure. Connectors scale automatically in response to increases in usage and you pay only for the resources you use. With full compatibility with Kafka Connect, it is easy to migrate workloads without code changes. MSK Connect will support both Amazon MSK-managed and self-managed Apache Kafka clusters.

You can get started with MSK Connect from the Amazon MSK console or the Amazon CLI. Visit the AWS Regions page for all the regions where Amazon MSK is available. To get started visit, the MSK Connect product page, pricing page, and the Amazon MSK Developer Guide.

 

​Amazon MSK Connect is now available in the Asia Pacific (Malaysia) Region. MSK Connect enables you to run fully managed Kafka Connect clusters with Amazon Managed Streaming for Apache Kafka (Amazon MSK). With a few clicks, MSK Connect allows you to easily deploy, monitor, and scale connectors that move data in and out of Apache Kafka and Amazon MSK clusters from external systems such as databases, file systems, and search indices. MSK Connect eliminates the need to provision and maintain cluster infrastructure. Connectors scale automatically in response to increases in usage and you pay only for the resources you use. With full compatibility with Kafka Connect, it is easy to migrate workloads without code changes. MSK Connect will support both Amazon MSK-managed and self-managed Apache Kafka clusters. You can get started with MSK Connect from the Amazon MSK console or the Amazon CLI. Visit the AWS Regions page for all the regions where Amazon MSK is available. To get started visit, the MSK Connect product page, pricing page, and the Amazon MSK Developer Guide.  

Publicado el — Deja un comentario

Amazon Q in Connect now supports selecting LLMs directly in Connect Web UI

Amazon Q in Connect, a generative AI-powered assistant for customer service, now enables contact center administrators to select different Large Language Models (LLMs) directly through the Amazon Connect web UI, providing a seamless AI Agent configuration experience. This no-code approach allows administrators to choose between LLM model families when building AI Agents to optimize for different business requirements. For example, you can select Amazon Nova Pro for faster response times, Anthropic Claude Sonnet for complex reasoning tasks, or switch between model families to optimize for different customer interaction types.

For more information about the AWS Regions where Amazon Q in Connect is available, see the Amazon Connect features by Region documentation. To learn more about Amazon Q in Connect, please visit the website or see the help documentation. To learn more about Amazon Connect, the AWS contact center as a service solution on the cloud, please visit the Amazon Connect website.

 

​Amazon Q in Connect, a generative AI-powered assistant for customer service, now enables contact center administrators to select different Large Language Models (LLMs) directly through the Amazon Connect web UI, providing a seamless AI Agent configuration experience. This no-code approach allows administrators to choose between LLM model families when building AI Agents to optimize for different business requirements. For example, you can select Amazon Nova Pro for faster response times, Anthropic Claude Sonnet for complex reasoning tasks, or switch between model families to optimize for different customer interaction types. For more information about the AWS Regions where Amazon Q in Connect is available, see the Amazon Connect features by Region documentation. To learn more about Amazon Q in Connect, please visit the website or see the help documentation. To learn more about Amazon Connect, the AWS contact center as a service solution on the cloud, please visit the Amazon Connect website.  

Publicado el — Deja un comentario

AWS Managed Microsoft AD adds LDAPS and Smart Card support using AWS Private CA

AWS Directory Service for Microsoft Active Directory (AWS Managed Microsoft AD) now offers certificate auto-enrollment for LDAPS and Smart Card and certificate based authentication with AWS Private Certificate Authority (AWS Private CA) through AWS Private CA Connector for AD. This integration enables automatic issuance, renewal, and management of certificates to AWS Managed Microsoft AD domain controllers, eliminating the need to maintain certificate authorities on Amazon EC2 instances.

By leveraging this fully managed solution, you can reduce costs of operating certificate authority infrastructure for Active Directory and simplify certificate management with AWS Private CA’s highly available, HSM-backed infrastructure. The integration supports LDAPS and smart card authentication while providing automatic certificate lifecycle management, flexible certificate control, and built-in security capabilities that streamline migration of Active Directory-aware workloads to AWS.

This feature is available in all AWS Regions where AWS Private CA Connector for AD is offered.

You can easily set up AWS Private CA integration with your directory in just a few clicks or programmatically via API. To get started, follow the step-by-step instructions in the Set up AWS Private CA Connector for AD for AWS Managed Microsoft AD documentation.

 

​AWS Directory Service for Microsoft Active Directory (AWS Managed Microsoft AD) now offers certificate auto-enrollment for LDAPS and Smart Card and certificate based authentication with AWS Private Certificate Authority (AWS Private CA) through AWS Private CA Connector for AD. This integration enables automatic issuance, renewal, and management of certificates to AWS Managed Microsoft AD domain controllers, eliminating the need to maintain certificate authorities on Amazon EC2 instances. By leveraging this fully managed solution, you can reduce costs of operating certificate authority infrastructure for Active Directory and simplify certificate management with AWS Private CA’s highly available, HSM-backed infrastructure. The integration supports LDAPS and smart card authentication while providing automatic certificate lifecycle management, flexible certificate control, and built-in security capabilities that streamline migration of Active Directory-aware workloads to AWS. This feature is available in all AWS Regions where AWS Private CA Connector for AD is offered. You can easily set up AWS Private CA integration with your directory in just a few clicks or programmatically via API. To get started, follow the step-by-step instructions in the Set up AWS Private CA Connector for AD for AWS Managed Microsoft AD documentation.  

Publicado el — Deja un comentario

Amazon EC2 R8g instances now available in additional regions

Starting today, Amazon Elastic Compute Cloud (Amazon EC2) R8g instances are available in AWS Asia Pacific (Osaka) and AWS Canada (Central) regions. These instances are powered by AWS Graviton4 processors and deliver up to 30% better performance compared to AWS Graviton3-based instances. Amazon EC2 R8g instances are ideal for memory-intensive workloads such as databases, in-memory caches, and real-time big data analytics. These instances are built on the AWS Nitro System, which offloads CPU virtualization, storage, and networking functions to dedicated hardware and software to enhance the performance and security of your workloads.

AWS Graviton4-based Amazon EC2 instances deliver the best performance and energy efficiency for a broad range of workloads running on Amazon EC2. AWS Graviton4-based R8g instances offer larger instance sizes with up to 3x more vCPU (up to 48xlarge) and memory (up to 1.5TB) than Graviton3-based R7g instances. These instances are up to 30% faster for web applications, 40% faster for databases, and 45% faster for large Java applications compared to AWS Graviton3-based R7g instances. R8g instances are available in 12 different instance sizes, including two bare metal sizes. They offer up to 50 Gbps enhanced networking bandwidth and up to 40 Gbps of bandwidth to the Amazon Elastic Block Store (Amazon EBS).

To learn more, see Amazon EC2 R8g Instances. To explore how to migrate your workloads to Graviton-based instances, see AWS Graviton Fast Start program and Porting Advisor for Graviton. To get started, see the AWS Management Console.

 

​Starting today, Amazon Elastic Compute Cloud (Amazon EC2) R8g instances are available in AWS Asia Pacific (Osaka) and AWS Canada (Central) regions. These instances are powered by AWS Graviton4 processors and deliver up to 30% better performance compared to AWS Graviton3-based instances. Amazon EC2 R8g instances are ideal for memory-intensive workloads such as databases, in-memory caches, and real-time big data analytics. These instances are built on the AWS Nitro System, which offloads CPU virtualization, storage, and networking functions to dedicated hardware and software to enhance the performance and security of your workloads. AWS Graviton4-based Amazon EC2 instances deliver the best performance and energy efficiency for a broad range of workloads running on Amazon EC2. AWS Graviton4-based R8g instances offer larger instance sizes with up to 3x more vCPU (up to 48xlarge) and memory (up to 1.5TB) than Graviton3-based R7g instances. These instances are up to 30% faster for web applications, 40% faster for databases, and 45% faster for large Java applications compared to AWS Graviton3-based R7g instances. R8g instances are available in 12 different instance sizes, including two bare metal sizes. They offer up to 50 Gbps enhanced networking bandwidth and up to 40 Gbps of bandwidth to the Amazon Elastic Block Store (Amazon EBS). To learn more, see Amazon EC2 R8g Instances. To explore how to migrate your workloads to Graviton-based instances, see AWS Graviton Fast Start program and Porting Advisor for Graviton. To get started, see the AWS Management Console.