Publicado el Deja un comentario

Amazon EventBridge Scheduler adds 619 new SDK API actions, including Lambda Managed Instances

Amazon EventBridge Scheduler expands its AWS SDK integrations with 13 additional services and 619 new API actions across new and existing AWS services, including AWS Lambda Managed Instances. You can now schedule direct invocations of a broader set of AWS services without writing custom integration code.

EventBridge Scheduler is a serverless scheduler that allows you to create, run, and manage billions of scheduled events and tasks across more than 270 AWS services, without provisioning or managing the underlying infrastructure. With this expansion, you can now schedule a broader set of AWS API actions directly from Scheduler, including scaling Lambda managed instances up or down on a time-based schedule for precise control over capacity provisioning.

These enhancements are now generally available in all AWS Regions where AWS EventBridge Scheduler is available. Specific services and API actions are subject to the availability of the target service in the AWS Region. To learn more about AWS EventBridge Scheduler SDK integrations, visit the Developer Guide.

 

​Amazon EventBridge Scheduler expands its AWS SDK integrations with 13 additional services and 619 new API actions across new and existing AWS services, including AWS Lambda Managed Instances. You can now schedule direct invocations of a broader set of AWS services without writing custom integration code. EventBridge Scheduler is a serverless scheduler that allows you to create, run, and manage billions of scheduled events and tasks across more than 270 AWS services, without provisioning or managing the underlying infrastructure. With this expansion, you can now schedule a broader set of AWS API actions directly from Scheduler, including scaling Lambda managed instances up or down on a time-based schedule for precise control over capacity provisioning. These enhancements are now generally available in all AWS Regions where AWS EventBridge Scheduler is available. Specific services and API actions are subject to the availability of the target service in the AWS Region. To learn more about AWS EventBridge Scheduler SDK integrations, visit the Developer Guide.  

Publicado el Deja un comentario

Amazon SageMaker Feature Store now supports SageMaker Python SDK V3

Amazon SageMaker Feature Store now supports the SageMaker Python SDK v3, including new capabilities for Lake Formation access controls and Apache Iceberg table properties configuration. Feature Store is a fully managed repository to store, share, and manage features for machine learning models. Data scientists can now use the modern, modular SDK v3 interfaces to manage feature groups with fine-grained access control and optimized offline storage.

Data scientists can use the SageMaker Python SDK v3 to manage feature groups with streamlined workflows and reduced boilerplate. With Lake Formation integration, data scientists can enforce column-level and row-level access control on offline store data through an opt-in setting at feature group creation. With Iceberg properties support, data scientists can configure additional table properties such as compaction and snapshot expiration directly through the SDK to optimize storage and query performance. These capabilities allow data scientists to govern access to feature data and optimize offline store performance from a single SDK without managing separate tools.

These capabilities are available in all AWS Regions where Amazon SageMaker Feature Store is available. To get started, install SageMaker Python SDK v3.8.0 or later. For more information, see Lake Formation access controls and Iceberg metadata management documentation.

 

​Amazon SageMaker Feature Store now supports the SageMaker Python SDK v3, including new capabilities for Lake Formation access controls and Apache Iceberg table properties configuration. Feature Store is a fully managed repository to store, share, and manage features for machine learning models. Data scientists can now use the modern, modular SDK v3 interfaces to manage feature groups with fine-grained access control and optimized offline storage. Data scientists can use the SageMaker Python SDK v3 to manage feature groups with streamlined workflows and reduced boilerplate. With Lake Formation integration, data scientists can enforce column-level and row-level access control on offline store data through an opt-in setting at feature group creation. With Iceberg properties support, data scientists can configure additional table properties such as compaction and snapshot expiration directly through the SDK to optimize storage and query performance. These capabilities allow data scientists to govern access to feature data and optimize offline store performance from a single SDK without managing separate tools. These capabilities are available in all AWS Regions where Amazon SageMaker Feature Store is available. To get started, install SageMaker Python SDK v3.8.0 or later. For more information, see Lake Formation access controls and Iceberg metadata management documentation.  

Publicado el Deja un comentario

Karpenter now supports Amazon Application Recovery Controller zonal shift

Amazon Elastic Kubernetes Service (Amazon EKS) now supports Amazon Application Recovery Controller (ARC) zonal shift and zonal autoshift when using the open source Karpenter project for compute provisioning. ARC helps you manage and coordinate recovery for your applications across AWS Regions and Availability Zones (AZs). With this launch, you can better maintain Kubernetes application availability by automating the process of shifting in-cluster network traffic away from an impaired AZ.

Customers increasingly deploy highly available applications in Amazon EKS across multiple AZs to eliminate a single point of failure. With ARC zonal shift, you can temporarily mitigate an AZ impairment by redirecting in-cluster network traffic away from the impacted AZ. For a fully automated experience, authorize AWS to manage this on your behalf using ARC zonal autoshift, which includes practice runs to verify your cluster functions as expected with one less AZ. When a zonal shift is activated for your EKS cluster, Karpenter stops provisioning new capacity in the impaired AZ, halts voluntary disruptions such as consolidation and drift for nodes in that AZ, and prevents voluntary disruptions in healthy zones if they depend on scheduling pods to the impaired zone. Pods with strict scheduling requirements such as volume affinities that require the impaired zone will not trigger launch attempts. When the zonal shift expires or is canceled, Karpenter resumes normal operations.

This Karpenter feature works with both manual zonal shifts and zonal autoshifts. No custom ARC resources are required as Karpenter integrates directly with the existing EKS cluster ARC resource. To enable zonal shift support, set the ENABLE_ZONAL_SHIFT setting in your Karpenter settings. To learn more, visit the Karpenter documentation and the ARC zonal shift documentation.

 

​Amazon Elastic Kubernetes Service (Amazon EKS) now supports Amazon Application Recovery Controller (ARC) zonal shift and zonal autoshift when using the open source Karpenter project for compute provisioning. ARC helps you manage and coordinate recovery for your applications across AWS Regions and Availability Zones (AZs). With this launch, you can better maintain Kubernetes application availability by automating the process of shifting in-cluster network traffic away from an impaired AZ. Customers increasingly deploy highly available applications in Amazon EKS across multiple AZs to eliminate a single point of failure. With ARC zonal shift, you can temporarily mitigate an AZ impairment by redirecting in-cluster network traffic away from the impacted AZ. For a fully automated experience, authorize AWS to manage this on your behalf using ARC zonal autoshift, which includes practice runs to verify your cluster functions as expected with one less AZ. When a zonal shift is activated for your EKS cluster, Karpenter stops provisioning new capacity in the impaired AZ, halts voluntary disruptions such as consolidation and drift for nodes in that AZ, and prevents voluntary disruptions in healthy zones if they depend on scheduling pods to the impaired zone. Pods with strict scheduling requirements such as volume affinities that require the impaired zone will not trigger launch attempts. When the zonal shift expires or is canceled, Karpenter resumes normal operations. This Karpenter feature works with both manual zonal shifts and zonal autoshifts. No custom ARC resources are required as Karpenter integrates directly with the existing EKS cluster ARC resource. To enable zonal shift support, set the ENABLE_ZONAL_SHIFT setting in your Karpenter settings. To learn more, visit the Karpenter documentation and the ARC zonal shift documentation.  

Publicado el Deja un comentario

Amazon Redshift launches RG instances powered by AWS Graviton

Amazon Redshift announces the general availability of RG instances, a new generation of provisioned cluster nodes powered by AWS Graviton processors that deliver better performance, running data warehouse and data lake workloads up to 2.4x as fast as previous generation RA3 instances, at 30% lower price per vCPU. RG instances include Redshift’s custom-built vectorized data lake query engine that processes Apache Iceberg and Parquet data on your cluster nodes — enabling you to run SQL analytics across your data warehouse and data lake using a single engine. This eliminates the need for Redshift Spectrum’s separate scanning fleet and its associated per-terabyte charges.

Whether you’re running structured data warehouse workloads on Redshift Managed Storage or querying open-format data lake tables in Amazon S3, RG instances deliver significant performance improvements — up to 2.2x as fast as RA3 instances for data warehouse workloads, up to 2.4x as fast for Apache Iceberg queries, and up to 1.5x as fast for Parquet workloads. The natively built data lake engine features a purpose-built I/O subsystem with smart prefetch, NVMe caching, vectorized Parquet scans, and advanced file and partition-level pruning. Just-in-Time (JIT) Analyze delivers consistently fast queries without manual tuning — automatically collecting and updating table statistics as your data and workload patterns evolve. Intelligent NVMe caching keeps frequently accessed datasets close to compute, reducing round-trips to your data lake for faster response times on repeated queries. RG instances are available at launch in two instance sizes — rg.xlarge and rg.4xlarge. Existing RA3 clusters can migrate using Snapshot & Restore, Elastic Resize, or Classic Resize. RG instances are available with flexible pricing options, including On-Demand, and 1-year and 3-year Reserved Instances with No Upfront payment. For pricing details, visit the Amazon Redshift pricing page.

Amazon Redshift RG instances are now available in the following AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), US West (N. California), Canada (Central), South America (São Paulo), Europe (Ireland), Europe (Frankfurt), Europe (London), Europe (Paris), Europe (Stockholm), Europe (Milan), Europe (Spain), Asia Pacific (Tokyo), Asia Pacific (Seoul), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Mumbai), Asia Pacific (Jakarta), Asia Pacific (Hong Kong), Asia Pacific (Osaka), Asia Pacific (Malaysia), Asia Pacific (Hyderabad), Asia Pacific (Taiwan), and Asia Pacific (Melbourne).

To get started, refer to the following resources:

 

​Amazon Redshift announces the general availability of RG instances, a new generation of provisioned cluster nodes powered by AWS Graviton processors that deliver better performance, running data warehouse and data lake workloads up to 2.4x as fast as previous generation RA3 instances, at 30% lower price per vCPU. RG instances include Redshift’s custom-built vectorized data lake query engine that processes Apache Iceberg and Parquet data on your cluster nodes — enabling you to run SQL analytics across your data warehouse and data lake using a single engine. This eliminates the need for Redshift Spectrum’s separate scanning fleet and its associated per-terabyte charges. Whether you’re running structured data warehouse workloads on Redshift Managed Storage or querying open-format data lake tables in Amazon S3, RG instances deliver significant performance improvements — up to 2.2x as fast as RA3 instances for data warehouse workloads, up to 2.4x as fast for Apache Iceberg queries, and up to 1.5x as fast for Parquet workloads. The natively built data lake engine features a purpose-built I/O subsystem with smart prefetch, NVMe caching, vectorized Parquet scans, and advanced file and partition-level pruning. Just-in-Time (JIT) Analyze delivers consistently fast queries without manual tuning — automatically collecting and updating table statistics as your data and workload patterns evolve. Intelligent NVMe caching keeps frequently accessed datasets close to compute, reducing round-trips to your data lake for faster response times on repeated queries. RG instances are available at launch in two instance sizes — rg.xlarge and rg.4xlarge. Existing RA3 clusters can migrate using Snapshot & Restore, Elastic Resize, or Classic Resize. RG instances are available with flexible pricing options, including On-Demand, and 1-year and 3-year Reserved Instances with No Upfront payment. For pricing details, visit the Amazon Redshift pricing page.
Amazon Redshift RG instances are now available in the following AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), US West (N. California), Canada (Central), South America (São Paulo), Europe (Ireland), Europe (Frankfurt), Europe (London), Europe (Paris), Europe (Stockholm), Europe (Milan), Europe (Spain), Asia Pacific (Tokyo), Asia Pacific (Seoul), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Mumbai), Asia Pacific (Jakarta), Asia Pacific (Hong Kong), Asia Pacific (Osaka), Asia Pacific (Malaysia), Asia Pacific (Hyderabad), Asia Pacific (Taiwan), and Asia Pacific (Melbourne).
To get started, refer to the following resources:

Amazon Redshift RG Instance Documentation
RA3 to RG Upgrade Guide
Amazon Redshift Pricing  

Publicado el Deja un comentario

De la capacidad a la responsabilidad: Asegurar nuestro ecosistema digital global con IA de próxima generación

De la capacidad a la responsabilidad: Asegurar nuestro ecosistema digital global con IA de próxima generación

Ilustración digital de un candado brillante sobre una red de líneas azules y puntos de datos, que representa ciberseguridad, cifrado y protección de la información.

Por: Amy Hogan-Burney, CVP, Seguridad y Confianza del Cliente.

La ciberseguridad está en un punto de inflexión. Los modelos avanzados de IA aceleran de manera importante el descubrimiento de vulnerabilidades y crean condiciones propicias para su explotación, todo esto subrayado por el anuncio de Claude Mythos Preview. Esto supone un cambio, y si esta tecnología favorecerá a los defensores o a los atacantes dependerá de las decisiones que tomemos ahora. 

Con las salvaguardas adecuadas, estas capacidades pueden ayudar a defensores de confianza a identificar y corregir vulnerabilidades en sistemas críticos en hospitales, redes eléctricas, agua y telecomunicaciones. Sin embargo, si se liberan de manera irresponsable o no están aseguradas de manera debida, esas mismas capacidades podrían ser abusadas por actores maliciosos, lo que podría amenazar los cimientos de nuestro ecosistema digital.

Gran parte del debate se ha centrado, con razón, en los riesgos. A medida que los modelos avanzados de IA aceleran el descubrimiento de vulnerabilidades, la manera en que las corregimos también debe acelerarse. Eso significa evaluaciones de riesgos previas al despliegue más sólidas y una estrecha colaboración entre gobiernos, desarrolladores de IA de vanguardia, proveedores de software y el ecosistema en general, para garantizar que estas herramientas reduzcan, en lugar de aumentar, el riesgo cibernético. Esto es en especial importante dado que los propios sistemas de IA se han convertido en objetivos de alto valor, lo que requiere una mayor protección de modelos, sistemas, datos e infraestructuras subyacentes.

Esto es, en última instancia, un desafío internacional. Ni las cadenas de suministro de software ni los actores amenazantes se detienen en las fronteras. Tampoco nuestra respuesta. Afrontar este momento requerirá enfoques compartidos entre países, sectores y sistemas, basados en la confianza, los estándares compartidos, la resiliencia y el uso responsable.

Este momento también es una oportunidad. La seguridad ha sido y sigue como la máxima prioridad en Microsoft. En los últimos dos años, a través de nuestra Iniciativa Futuro Seguro, hemos reforzado nuestras bases de seguridad para esta era de la IA, en parte al utilizar la IA para acelerar el descubrimiento y la remediación de vulnerabilidades. También hemos invertido en IA fundamental para la investigación en seguridad, incluido el desarrollo de benchmarks industriales de código abierto que puedan utilizarse para evaluar si los modelos están preparados para trabajos reales de seguridad. Aceleramos ese trabajo mediante una colaboración público-privada más profunda y en colaboración con IA, incluido el Proyecto Glasswing de Anthropic y el programa Trusted Access for Cyber de OpenAI. 

Proteger nuestro ecosistema digital con IA de próxima generación está al alcance, pero no es automático.

Construir bases seguras para la era de la IA Frontier

Garantizar que las tecnologías avanzadas de IA se utilicen para fortalecer la ciberseguridad requiere una acción deliberada y urgente. Compartimos las siguientes recomendaciones como medidas prácticas que gobiernos, industria y el ecosistema en general pueden tomar para garantizar que estas herramientas, a menudo denominadas «IA Frontier», refuercen los cimientos de seguridad de los que dependen las sociedades digitales. Y esperamos seguir con la colaboración con proveedores modelo, la industria y el gobierno para poder trabajar juntos y mejorar los resultados de seguridad para todos.

1. Reforzar las prácticas básicas de ciberseguridad

La IA avanzada solo puede fortalecer la ciberseguridad cuando ya existe una higiene cibernética fuerte y constante. A medida que la IA pionera acelera el descubrimiento y la respuesta a vulnerabilidades, prácticas clave como el parche rápido, el control de accesos y la resiliencia del sistema se vuelven más críticas, no menos.

Los avances en seguridad en la era pionera de la IA dependen de la estrecha coordinación entre los proveedores tecnológicos que avanzan en nuevas capacidades y las organizaciones responsables de operar, actualizar y asegurar sistemas reales. Sin esta interdependencia, la IA avanzada no puede ofrecer mejoras duraderas en seguridad. Ninguna organización puede resolver estos problemas de ciberseguridad por sí sola.

Por eso, la inversión sostenida en lo que sabemos que funciona es todavía esencial: ciclos de vida de productos de secure-by-design (seguro desde el diseño), arquitecturas Zero Trust (Confianza Cero), autenticación multifactor, acceso menos privilegiado y formación continua en seguridad. Adopción y armonización generalizada de marcos de ciberseguridad establecidos para garantizar una resiliencia coherente en los sistemas habilitados por IA. Entornos de nube de confianza que permiten estas prácticas a gran escala, al apoyar el manejo seguro de datos, el parche continuo y el despliegue seguro de herramientas habilitadas por IA para defensores.

2. Liberar capacidades avanzadas de manera responsable

A medida que los sistemas de IA de vanguardia adquieren capacidades de razonamiento, codificación y agentes, surgen algunos de los riesgos de seguridad más graves antes del despliegue, incluido un uso indebido realista que implica razonamiento en varios pasos, uso de herramientas y reconocimiento. Los benchmarks técnicos de seguridad son todavía importantes, pero son insuficientes sin pruebas rigurosas y realistas.

Como resultado, los gobiernos establecen cada vez más evaluaciones previas al despliegue que combinan pruebas técnicas con modelado de amenazas. Estas evaluaciones son más efectivas cuando los Desarrolladores Frontier trabajan de manera estrecha con organizaciones que monitorizan los riesgos de seguridad nacional. Invertir en entornos de evaluación seguros y métodos modernos de prueba puede ayudar a los gobiernos a mantenerse al día a medida que avanzan las capacidades.  

Las prácticas de liberación responsable, incluido el acceso por fases y controlado, son una extensión crítica de este enfoque. Nuestro trabajo con Anthropic en el Proyecto Glasswing ofrece un modelo práctico, que permite a defensores de confianza evaluar capacidades avanzadas en entornos restringidos antes de su lanzamiento más amplio. De manera similar, OpenAI y Microsoft trabajan de manera estrecha a través del programa Trusted Access for Cyber, y ya apoyamos el uso de despliegues tempranos y con alcance definido para pruebas de seguridad y protección.  

La responsabilidad no termina en la liberación. Las organizaciones que implementan modelos Frontier suelen estar en mejor posición para detectar nuevos abusos y deben monitorizar, mitigar y compartir información sobre amenazas. Microsoft colabora con sus pares a través del Frontier Model Forum para avanzar en las mejores prácticas en la evaluación y gestión del riesgo cibernético y facilitar el intercambio de información. Los gobiernos deberían fomentar la colaboración continua con la industria para restringir el acceso de actores amenazantes identificados y contrarrestar el uso adversarial o malicioso de IA avanzada.

3. Modernizar la gestión de vulnerabilidades

La IA cambia tanto la velocidad de detección de vulnerabilidades como lo que constituye un riesgo de seguridad significativo. Un descubrimiento más rápido solo mejora la seguridad si el triaje, la validación y la remediación pueden mantenerse al día.

A medida que la IA acelera el descubrimiento, la gestión de vulnerabilidades debe pasar de rastrear el volumen bruto a reducir el riesgo en el mundo real. Eso significa priorizar vulnerabilidades que sean en verdad explotables, asignar una responsabilidad clara en el triaje y la remediación, y utilizar una divulgación por fases y basada en riesgos cuando la coordinación privada mejore la seguridad. Por encima de todo, los sistemas deben diseñarse en torno a la validación y la capacidad realista de remediación, no a la suposición de que más hallazgos conducen en automático a una mejor seguridad.

Los desarrolladores de modelos de IA vanguardistas deberían integrar la coordinación y divulgación de vulnerabilidades directo en los marcos de redacción responsable. Y trabajar con gobiernos e industria para asegurar que los hallazgos se dirijan a los propietarios adecuados, se actúe con antelación y se respalden mediante vías claras de coordinación.

4. Arreglar más rápido: Fortalecer y acelerar la respuesta y la remediación

A medida que la IA acelera el descubrimiento de vulnerabilidades, la remediación debe mantener el ritmo. Iniciativas como el AI Cyber Challenge de DARPA demuestran cómo la IA puede ayudar tanto a encontrar como a corregir fallos en el software de código abierto. Reforzar las defensas requiere inversión no solo en herramientas de detección, sino también en las personas, procesos e infraestructuras responsables de corregir vulnerabilidades, en especial en sectores críticos. 

Gran parte del software que sustenta la infraestructura crítica depende de componentes de código abierto mantenidos por pequeños equipos o voluntarios con capacidad de seguridad limitada. Un aumento en el descubrimiento habilitado por IA corre el riesgo de sobrepasar los procesos de triaje y divulgación existentes. Iniciativas como el GitHub Secure Open Source Fund, junto con inversiones de Microsoft y otros a través de la Linux Foundation, AlphaOmega y OpenSSF, ayudan a los mantenedores a adaptarse de manera práctica y alineada con los flujos de trabajo existentes.  

Los gobiernos deberían tratar la capacidad de remediación como una prioridad fundamental en la resiliencia, incluida la inversión sostenida y el apoyo a los mantenedores, la capacidad de aumento durante grandes eventos de descubrimiento y la modernización de las vías de divulgación, para reconocer que la remediación eficaz aun depende en gran medida del juicio humano, la coordinación y el tiempo.

5. Avanzar en la seguridad de la IA a nivel internacional

La seguridad de la IA es esencial para desplegar la IA a gran escala. Dado que los sistemas de IA, las cadenas de suministro y los riesgos que introducen operan a través de las fronteras, los enfoques nacionales por sí solos no serán suficientes.

Los gobiernos y la industria deberían trabajar juntos para construir bases internacionales interoperables para la seguridad de la IA, incluida la evaluación de riesgos, la divulgación coordinada de vulnerabilidades y el intercambio de información. Las prioridades deberían incluir fortalecer el uso defensivo de la IA, prevenir el mal uso mediante normas y salvaguardas compartidas, y asegurar los sistemas de IA y la pila tecnológica de IA.

La participación global es fundamental. Los países y organizaciones con recursos limitados de ciberseguridad o infraestructuras heredadas suelen ser los más expuestos. La cooperación internacional debe priorizar el fortalecimiento de capacidades, para asegurar que los beneficios de seguridad de la IA se realicen de manera amplia y equitativa.

La seguridad de la IA no es solo una salvaguarda; es un facilitador de innovación y crecimiento. Al actuar de manera colectiva y moverse con rapidez, los gobiernos y la industria pueden fortalecer la resiliencia digital global y desbloquear la adopción confiable de la IA en economías, infraestructuras críticas y servicios públicos.

Afrontar el momento: Utilizar capacidades de IA de vanguardia para generar confianza y confianza

Afrontar este momento es, en última instancia, una cuestión de confianza: no en una tecnología o proveedor individual, sino en nuestra capacidad colectiva para introducir IA avanzada de manera responsable.

Utilizadas de forma deliberada y basadas en sólidas bases de seguridad, estas capacidades pueden reforzar la ciberseguridad y reforzar la confianza en los sistemas de los que depende la sociedad. La elección no es entre innovación y seguridad, sino si permitimos que se refuercen de manera mutua.

Ese resultado está al alcance. Con gobiernos, industria y operadores de infraestructuras alineados, la IA avanzada puede desplegarse de formas que coincidan con la capacidad defensiva real y apoyen acciones legales y de confianza. Si se hace bien y se trabaja en conjunto, la IA de vanguardia puede ayudar a proteger la infraestructura digital que sustenta la vida moderna y generar una confianza duradera en su resiliencia.

The post De la capacidad a la responsabilidad: Asegurar nuestro ecosistema digital global con IA de próxima generación appeared first on Source LATAM.

 

​The post De la capacidad a la responsabilidad: Asegurar nuestro ecosistema digital global con IA de próxima generación appeared first on Source LATAM.  

Publicado el Deja un comentario

Announcing Region Expansion of P6-B200 instances on SageMaker Studio notebooks

We are pleased to announce general availability of Amazon EC2 P6-B200 instances in AWS US East (N. Virginia) on SageMaker Studio notebooks.

Amazon EC2 P6-B200 instances are powered by 8 NVIDIA Blackwell GPUs with 1440 GB of high-bandwidth GPU memory and 5th Generation Intel Xeon processors (Emerald Rapids). These instances deliver up to 2x better performance compared to P5en instances for AI training. Customers can use P6-B200 instances to interactively develop and fine-tune large foundation models, including LLMs, mixture of experts models, and multi-modal reasoning models. These instances enable efficient experimentation with larger models directly in JupyterLab or CodeEditor environments for generative AI applications such as enterprise copilots and content generation across text, images, and video.

Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.

 

​We are pleased to announce general availability of Amazon EC2 P6-B200 instances in AWS US East (N. Virginia) on SageMaker Studio notebooks.
Amazon EC2 P6-B200 instances are powered by 8 NVIDIA Blackwell GPUs with 1440 GB of high-bandwidth GPU memory and 5th Generation Intel Xeon processors (Emerald Rapids). These instances deliver up to 2x better performance compared to P5en instances for AI training. Customers can use P6-B200 instances to interactively develop and fine-tune large foundation models, including LLMs, mixture of experts models, and multi-modal reasoning models. These instances enable efficient experimentation with larger models directly in JupyterLab or CodeEditor environments for generative AI applications such as enterprise copilots and content generation across text, images, and video.
Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.  

Publicado el Deja un comentario

ENA Express for Amazon EC2 instances now supports traffic between Availability Zones

Elastic Network Adapter (ENA) Express now supports traffic between Amazon EC2 instances in different Availability Zones within a Region, delivering up to 25 Gbps single-flow bandwidth. ENA Express is a networking feature that uses the AWS Scalable Reliable Datagram (SRD) protocol to improve network performance. SRD is a reliable network protocol that delivers performance improvements through advanced congestion control and multi-pathing. Amazon Elastic Block Store (EBS) io2 Block Express and Elastic Fabric Adapter (EFA) for high performance computing and machine learning workloads also leverage SRD.

Workloads such as distributed storage, databases, and file systems require deployments spanning multiple Availability Zones for resilience, yet single flows between zones support up to 5 Gbps with ENA. ENA Express delivers up to 25 Gbps single-flow bandwidth for traffic between Availability Zones. To achieve this, ENA Express detects compatibility between your EC2 instances and establishes an SRD connection when both communicating instances have ENA Express enabled. Once established, SRD uses multi-pathing to route your traffic across the network and avoids head-of-line blocking as it does not need packets to arrive in order. Using these capabilities, ENA Express delivers the performance benefits transparently to your application with TCP and UDP protocols.

ENA Express for connections between Availability Zones within a Region is available for all supported instance types and sizes in Africa (Cape Town), Asia Pacific (Hong Kong, Hyderabad, Jakarta, Malaysia, Melbourne, Mumbai, New Zealand, Osaka, Seoul, Singapore, Sydney, Taipei, Thailand, Tokyo), Canada (Central), Canada West (Calgary), Europe (Frankfurt, Ireland, London, Milan, Paris, Spain, Stockholm, Zurich), Israel (Tel Aviv), Mexico (Central), US East (N. Virginia, Ohio), US West (N. California, Oregon), and AWS GovCloud (US) Regions. ENA Express comes at no additional cost. For a list of supported instances and configuration guidance, please review the latest EC2 documentation.

 

​Elastic Network Adapter (ENA) Express now supports traffic between Amazon EC2 instances in different Availability Zones within a Region, delivering up to 25 Gbps single-flow bandwidth. ENA Express is a networking feature that uses the AWS Scalable Reliable Datagram (SRD) protocol to improve network performance. SRD is a reliable network protocol that delivers performance improvements through advanced congestion control and multi-pathing. Amazon Elastic Block Store (EBS) io2 Block Express and Elastic Fabric Adapter (EFA) for high performance computing and machine learning workloads also leverage SRD.
Workloads such as distributed storage, databases, and file systems require deployments spanning multiple Availability Zones for resilience, yet single flows between zones support up to 5 Gbps with ENA. ENA Express delivers up to 25 Gbps single-flow bandwidth for traffic between Availability Zones. To achieve this, ENA Express detects compatibility between your EC2 instances and establishes an SRD connection when both communicating instances have ENA Express enabled. Once established, SRD uses multi-pathing to route your traffic across the network and avoids head-of-line blocking as it does not need packets to arrive in order. Using these capabilities, ENA Express delivers the performance benefits transparently to your application with TCP and UDP protocols.
ENA Express for connections between Availability Zones within a Region is available for all supported instance types and sizes in Africa (Cape Town), Asia Pacific (Hong Kong, Hyderabad, Jakarta, Malaysia, Melbourne, Mumbai, New Zealand, Osaka, Seoul, Singapore, Sydney, Taipei, Thailand, Tokyo), Canada (Central), Canada West (Calgary), Europe (Frankfurt, Ireland, London, Milan, Paris, Spain, Stockholm, Zurich), Israel (Tel Aviv), Mexico (Central), US East (N. Virginia, Ohio), US West (N. California, Oregon), and AWS GovCloud (US) Regions. ENA Express comes at no additional cost. For a list of supported instances and configuration guidance, please review the latest EC2 documentation.  

Publicado el Deja un comentario

Announcing Region Expansion of G6e instances on SageMaker Studio notebooks

We are pleased to announce general availability of Amazon EC2 G6e instances in the Middle East (Dubai), Asia Pacific (Tokyo, Seoul) and Europe (Frankfurt, Stockholm, Spain) on SageMaker Studio notebooks.

Amazon EC2 G6e instances are powered by up to 8 NVIDIA L40s Tensor Core GPUs with 48 GB of memory per GPU and third generation AMD EPYC processors. G6e instances deliver up to 2.5x better performance compared to EC2 G5 instances. Customers can use G6e instances to interactively test model deployment and for interactive model training use cases such as generative AI fine-tuning. You can use G6e instances to deploy large language models (LLMs) with up to 13B parameters and diffusion models for generating images, video, and audio.

Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.

 

​We are pleased to announce general availability of Amazon EC2 G6e instances in the Middle East (Dubai), Asia Pacific (Tokyo, Seoul) and Europe (Frankfurt, Stockholm, Spain) on SageMaker Studio notebooks.
Amazon EC2 G6e instances are powered by up to 8 NVIDIA L40s Tensor Core GPUs with 48 GB of memory per GPU and third generation AMD EPYC processors. G6e instances deliver up to 2.5x better performance compared to EC2 G5 instances. Customers can use G6e instances to interactively test model deployment and for interactive model training use cases such as generative AI fine-tuning. You can use G6e instances to deploy large language models (LLMs) with up to 13B parameters and diffusion models for generating images, video, and audio.
Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.  

Publicado el Deja un comentario

Announcing Region Expansion of P4de instances on SageMaker Studio notebooks

We are pleased to announce general availability of Amazon EC2 P4de instances in Asia Pacific (Tokyo, Singapore) and Europe (Frankfurt) on SageMaker Studio notebooks.

Amazon EC2 P4de instances are powered by 8 NVIDIA A100 GPUs with 80GB high-performance HBM2e GPU memory, 2X higher than the GPUs in our current P4d instances. The new P4de instances provide a total of 640GB of GPU memory, which provide up to 60% better ML training performance along with 20% lower cost to train when compared to P4d instances. The improved performance will allow customers to reduce model training times and accelerate time to market. Increased GPU memory on P4de will also benefit workloads that need to train on large datasets of high-resolution data.

Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.

 

​We are pleased to announce general availability of Amazon EC2 P4de instances in Asia Pacific (Tokyo, Singapore) and Europe (Frankfurt) on SageMaker Studio notebooks. Amazon EC2 P4de instances are powered by 8 NVIDIA A100 GPUs with 80GB high-performance HBM2e GPU memory, 2X higher than the GPUs in our current P4d instances. The new P4de instances provide a total of 640GB of GPU memory, which provide up to 60% better ML training performance along with 20% lower cost to train when compared to P4d instances. The improved performance will allow customers to reduce model training times and accelerate time to market. Increased GPU memory on P4de will also benefit workloads that need to train on large datasets of high-resolution data. Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.  

Publicado el Deja un comentario

Announcing Region Expansion of G6 instances on SageMaker Studio notebooks

We are pleased to announce general availability of Amazon EC2 G6 instances in the Middle East (Dubai) and Asia Pacific (Malaysia) on SageMaker Studio notebooks.

Amazon EC2 G6 instances are powered by up to 8 NVIDIA L4 Tensor Core GPUs with 24 GB of memory per GPU and third generation AMD EPYC processors. G6 instances offer 2x better performance for deep learning inference compared to EC2 G4dn instances. Customers can use G6 instances to interactively test model deployment and for interactive model training for use cases such as generative AI fine-tuning and inference workloads, natural language processing, language translation, computer vision, and recommender engines.

Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.

 

​We are pleased to announce general availability of Amazon EC2 G6 instances in the Middle East (Dubai) and Asia Pacific (Malaysia) on SageMaker Studio notebooks.
Amazon EC2 G6 instances are powered by up to 8 NVIDIA L4 Tensor Core GPUs with 24 GB of memory per GPU and third generation AMD EPYC processors. G6 instances offer 2x better performance for deep learning inference compared to EC2 G4dn instances. Customers can use G6 instances to interactively test model deployment and for interactive model training for use cases such as generative AI fine-tuning and inference workloads, natural language processing, language translation, computer vision, and recommender engines.
Visit developer guides for instructions on setting up and using JupyterLab and CodeEditor applications on SageMaker Studio. For pricing information on these instances, please visit our pricing page.