AWS re:Invent 2020 Compute Announcements

AWS Categories: Compute

Amazon EC2 Mac Instances

What is it?

Mac instances enable customers to run on-demand macOS workloads in the cloud for the first time, extending the flexibility, scalability, and cost benefits of AWS to all Apple developers. Customers who rely on the Xcode IDE for creating iPhone, iPad, Mac, Apple Watch, Apple TV, and Safari apps can now provision and access macOS environments within minutes with simple mouse clicks or API calls, dynamically scale capacity as needed, and benefit from AWS’s pay-as-you-go pricing.

Amazon EC2 Mac instances are built on Mac minicomputers and offer customers a choice of both the macOS Mojave (10.14) and macOS Catalina (10.15) versions.

Availability:

us-east-1 (N. Virginia), us-east-2 (Ohio), us-west-2 (Oregon), eu-west-1 (Ireland), and ap-southeast-1 (Singapore)

Customer Benefits:

  • Quickly provision macOS environments: Time and resources previously spent building and maintaining on-premises macOS environments can now be refocused on building creative and useful apps. Development teams can now seamlessly provision and access macOS compute environments to enjoy faster app builds and convenient, distributed testing, without having to procure, configure, operate, maintain, and upgrade fleets of physical computers.
  • Reduce costs: Mac instances allow developers to launch macOS environments within minutes, adjust provisioned capacity as needed, and only pay for actual usage with AWS’s pay-as-you-go pricing. Developers save money since they only need to pay for the systems that are in use. For example, more capacity can be used when building an app, and less capacity when testing.
  • Extend your toolkits: Amazon EC2 Mac instances provide developers seamless access to the broad set of over 175 AWS services so they can more easily and efficiently collaborate with team members, and develop, test, share, analyze, and improve their apps. Customers can leverage AWS services such as Elastic Block Store (EBS) for block-level storage, Elastic Load Balancer (ELB) for distributing build queues, Simple Storage Service (S3) for extreme scale object storage, Amazon Machine Images (AMIs) for orchestration, and CodeBuild for managed CI/CD.

Resources: Website | What’s New Post

Amazon EC2 D3 and D3en Instances

What is it?

Amazon EC2 D3 and D3en instances provide cost-effective, high-capacity local storage-per-vCPU for massively scaled storage workloads. D3 and D3en instances are the next generation of dense HDD storage instances, offering 30% higher processor performance, increased capacity, and reduced cost compared to D2 instances. Additionally, D3 instances provide 2.5x higher networking speed and 45% higher disk throughput compared to D2 instances. D3en instances, enhanced storage and high-speed networking variants, provide 7.5x higher networking speed, 100% higher disk throughput, 7x more storage capacity (up to 336 TB), and 80% lower cost per-TB of storage compared to D2 instances.

D3 instances are a great fit for dense storage workloads including big data and analytics, data warehousing, and high scale file systems. D3en instances are a great fit for dense and distributed workloads including high-capacity data lakes, clustered file systems, and other multi-node storage systems with significant inter-node I/O. With D3 and D3en instances, you can easily migrate from previous- generation D2 instances or on-premises infrastructure to a platform optimized for dense HDD storage workloads.

Availability:

us-east-1, us-east-2, us-west-2, and eu-west-1 regions

Customer Benefits:

  • Lower costs: Next-generation Amazon EC2 D3 instances provide increased price-performance, and lower cost than D2 instances. D3 and D3en instances feature 30% higher compute performance than D2 instances. D3en instances also offer 80% lower cost-per-TB of storage compared to D2 instances.
  • Better performance: D3 and D3en instances satisfy the needs of applications with high requirements for sequential storage throughput. D3 and D3en instances enable 45% and 100% higher disk throughput respectively compared to D2 instances. D3 and D3en instances provide 2.5x and 7.5x higher networking throughput respectively than D2 instances, allowing for high-speed multi-node configurations.
  • Maximize resource efficiency: D3 and D3en instances are powered by the AWS Nitro System, a combination of dedicated hardware and lightweight hypervisor, which delivers practically all of the compute and memory resources of the host hardware to your instances. This frees up additional compute, memory and I/O, allowing your applications to do more with available hardware resources including local, HDD storage.

Resources: Website | What’s New Post

Amazon EC2 Instances Powered by AWS Graviton2 Processors

What is it?

The new general purpose (M6g), general purpose burstable (T4g), compute optimized (C6g), and memory optimized (R6g) Amazon EC2 instances deliver up to 40% improved price performance over comparable x86-based instances for a broad spectrum of workloads including application servers, open-source databases, in-memory caches, microservices, gaming servers, electronic design automation, high-performance computing, and video encoding. M6gd, C6gd, and R6gd are variants of these instances with local NVMe-based SSD storage, and C6gn instances deliver 100 Gbps networking for compute intensive applications with support for Elastic Fabric Adapter (EFA). These instances are powered by new AWS Graviton2 processors that deliver up to 7x performance, 4x the number of compute cores, 2x larger private caches per core, and 5x faster memory compared to the first-generation AWS Graviton Processors.

AWS Graviton2 processors are built on advanced 7 nanometer manufacturing technology. They utilize 64-bit Arm Neoverse cores and custom silicon designed by AWS and introduce several performance optimizations versus the first generation. AWS Graviton2 processors provide 2x faster floating-point performance per core for scientific and high-performance computing workloads, custom hardware acceleration for compression workloads, fully encrypted DRAM memory, and optimized instructions for faster CPU-based machine learning inference.

Availability:

US East (N. Virginia, Ohio), US West (N. California, Oregon), Europe (Ireland, Frankfurt, London), Canada (Central) and Asia Pacific (Mumbai, Singapore, Sydney, Tokyo) regions

Customer Benefits:

  • Best price performance for a broad spectrum of workloads: AWS Graviton2-based general-purpose (M6g), general-purpose burstable (T4g), compute-optimized (C6g), and memory-optimized (R6g) EC2 instances deliver up to 40% better price performance over comparable current generation x86-based instances for a broad spectrum of workloads such as application servers, micro-services, video encoding, high-performance computing, electronic design automation, compression, gaming, open-source databases, in-memory caches, and CPU-based machine learning inference.
  • Extensive ecosystem support: AWS Graviton2 processors, based on the 64-bit Arm architecture, are supported by popular Linux operating systems including Amazon Linux 2, Red Hat, SUSE, and Ubuntu. Many popular applications and services from AWS and Independent Software Vendors also support AWS Graviton2-based instances, including Amazon ECS, Amazon EKS, Amazon ECR, Amazon CodeBuild, Amazon CodeCommit, Amazon CodePipeline, Amazon CodeDeploy, Amazon CloudWatch, Crowdstrike, Datadog, Docker, Drone, GitLab, Jenkins, NGINX, Qualys, Rancher, Rapid7, Tenable, and TravisCI. Arm developers can also leverage this ecosystem to build applications natively in the cloud, thereby eliminating the need for emulation and cross-compilation, which are error prone and time consuming.
  • Enhanced security for cloud applications: Developers building applications for the cloud rely on cloud infrastructure for security, speed and optimal resource footprint. AWS Graviton2 processors feature key capabilities that enables developers to run cloud native applications securely, and at scale, including always-on 256-bit DRAM encryption and 50% faster per core encryption performance compared to first-generation AWS Graviton. Graviton2 powered instances are built on the Nitro System that features the Nitro security chip with dedicated hardware and software for security functions, as well as encrypted EBS storage volumes by default.

Resources: Website

Amazon EC2 G4ad instances

What is it?

G4ad instances are powered by AMD Radeon Pro V520 GPUs, providing the best price performance for graphics intensive applications in the cloud. These instances offer up to 45% better price performance compared to G4dn instances, which were already the lowest cost instances in the cloud, for graphics applications such as remote graphics workstations, game streaming, and rendering that leverage industry-standard APIs such as OpenGL, DirectX, and Vulkan. They provide up to 4 AMD Radeon Pro V520 GPUs, 64 vCPUs, 25 Gbps networking, and 2.4 TB local NVMe-based SSD storage.

Availability:

us-east-1 (N. Virginia), us-east-2 (Ohio), us-west-2 (Oregon), eu-west-1 (Ireland), and ap-southeast-1 (Singapore)

Use Cases:

  • Virtual Workstations
  • Graphics intensive applications

Customer Benefits:

  • Highest Performance and Lowest Cost Instances for Graphics Intensive Applications: G4ad instances are the lowest cost instances in the cloud for graphics intensive applications. They provide up to 45% better price performance, including up to 40% better graphics performance, compared to G4dn instances for graphics applications such as remote graphics workstations, game streaming, and rendering that leverage industry standard APIs such as OpenGL, DirectX, and Vulkan.
  • Simplified Management of Virtual Workstations at the Lowest Cost in the Cloud: G4ad instances allow customers to configure virtual workstations with high-performance simulation, rendering, and design capabilities in minutes, allowing customers to scale quickly. Customers can use AMD Radeon Pro Software for Enterprise and high-performance remote display protocol, NICE DCV, with G4ad instances at no additional cost to manage their virtual workstation environments with support for up to two 4k monitors per GPU.
  • Dependability in Third Party Applications: The AMD professional graphics solution includes an extensive Independent Software Vendor (ISV) application testing and certification process called the Day Zero Certification Program. This helps ensure that developers can leverage the latest AMD Radeon Pro Software for Enterprise features combined with the reliability of certified software on the day of the driver release.

Resources: Website

AWS Wavelength Zone in Las Vegas

What is it?

Today, we are announcing the availability of a new AWS Wavelength Zone on Verizon’s 5G Ultra-Wideband network in Las Vegas. Wavelength Zones are now available in eight cities, including the seven previously announced cities of Boston, San Francisco Bay Area, New York City, Washington DC, Atlanta, Dallas, and Miami.

AWS Wavelength brings AWS services to the edge of the 5G network, minimizing the latency to connect to an application from 5G connected devices. Application traffic can reach application servers running in Wavelength Zones, AWS infrastructure deployments that embed AWS compute and storage services within the communications service providers’ datacenters at the edge of the 5G networks, without leaving the telco provider’s network. This reduces the extra network hops to the Internet that can result in latencies of 10s of milliseconds, preventing customers from taking full advantage of the bandwidth and latency advancements of 5G.

Availability:

Today, Wavelength was announced for availability in Las Vegas. In August 2020, AWS announced the launch of two Wavelength Zones, in San Francisco and Boston, with Verizon. Wavelength Zones in 8 other cities in the United States are planned for launch in 2020. Globally, AWS is partnering with other leading-edge telecommunications companies including KDDI, SK Telecom, and Vodafone to launch Wavelength across Europe, Japan, and South Korea in 2020, with more telco partners coming soon.

Use cases:

  • Connected Vehicles: Cellular Vehicle-to-Everything (C-V2X) is an increasingly important platform for enabling intelligent driving, real-time HD-maps, road safety, and more.
  • Interactive Live Video Streams: Wavelength provides the ultra-low latency needed to live stream high-resolution video and high-fidelity audio, as well as to embed interactive experiences into live video streams.
  • AR/VR: By accessing compute resources on AWS Wavelength, AR/VR applications can reduce the Motion to Photon (MTP) latencies to the <20 ms benchmark needed to offer a realistic customer experience.
  • Smart Factories: Industrial automation applications use ML inference at the edge to analyze images and videos to detect quality issues on fast moving assembly lines and trigger actions to remediate the problem.
  • Real-time gaming: With AWS Wavelength, the most demanding games can be made available on end devices that have limited processing power by streaming these games from game servers in Wavelength Zones.
  • Healthcare: ML-assisted diagnostics: AI/ML driven video analytics and image matching solutions help doctors speed up diagnosis of observed conditions.

Customer Benefits:

  • Ultra-low latency for 5G: Wavelength combines AWS compute and storage services with the high bandwidth and low latency of 5G networks to enable developers to innovate and build a whole new class of applications that serve end-users with ultra-low latencies over the 5G network.
  • Consistent AWS experience: Wavelength enables you to use familiar and powerful AWS tools and services to build, manage, secure, and scale your applications.
  • Flexible and scalable: With Wavelength, you can start small and scale as your needs grow, without worrying about managing physical hardware or maximizing the utilization of purchased capacity.
  • Global 5G network: Wavelength will be available within communications service providers’ (CSP) networks such as Verizon, Vodafone, KDDI, and SK Telecom. More CSPs around the world will be available in the near future.

Resources: Website | What’s New Post

Amazon EC2 instances powered by Habana Accelerators

What is it?

Amazon EC2 instances powered by Habana accelerators are a new type of EC2 instance specifically optimized for deep learning training workloads to deliver the lowest cost-to-train machine learning models in the cloud. Habana-based instances are ideal for deep learning training workloads of applications such as natural language processing, object detection and classification, recommendation engines and autonomous vehicle perception. Habana, an Intel company, will provide the SynapseAI SDK and tools that simplify building with or migrating from current GPU-based EC2 instances to Habana-based EC2 instances.

SynapseAI will be natively integrated with common ML frameworks like TensorFlow and PyTorch and provide the ability to easily port existing training models from using GPUs to Habana accelerators. Customers will be able to launch the new EC2 instances using AWS Deep Learning AMIs, or via Amazon EKS and ECS for containerized applications, and also have the ability to use these instances via Amazon Sagemaker.

Availability:

Amazon EC2 Habana-based instances will be available in April 2021 in 3 sizes across in 2 regions: us-east-1 and us-west-2. They can be purchased as On-Demand, Reserved Instances, Savings Plan or Spot Instances. Habana-based instances are also available for use with Amazon SageMaker and Amazon EKS.

Customer Benefits:

  • Better performance and lower cost: Habana based EC2 instances will leverage up to 8 Habana Gaudi accelerators and deliver up to 40% better price performance than current GPU-based EC2 instances for training deep learning models. Habana-based instances also provide customers the ability to scale out from a single accelerator to hundreds in significantly reducing time-to-train.

Resources: Website

AWS Trainium

What is it?

AWS Trainium is high performance machine learning (ML) chip, custom designed by AWS to provide the best price performance for training machine learning models in the cloud. The Trainium chip is specifically optimized for deep learning training workloads for applications including image classification, semantic search, translation, voice recognition, natural language processing and recommendation engines. AWS Trainium uses the AWS Neuron SDK which is integrated with popular ML frameworks, including TensorFlow, MXNet, and PyTorch, allowing customers to easily migrate from using GPU instances for training deep learning models with minimal code changes. AWS Trainium will be available via Amazon EC2 instances and AWS Deep Learning AMIs as well as managed services including Amazon SageMaker, Amazon ECS, EKS, and AWS Batch.

Availability:

AWS Trainium will be available in all AWS commercial and GovCloud regions.

Use Cases:

  • Virtual Workstations
  • Graphics intensive applications

Customer Benefits:

  • Better performance and lower cost: AWS Trainium will deliver the most cost-effective ML training in the cloud and will offer the most TFLOPS of compute power of any ML instance in the cloud. Customers will be able to achieve significantly better performance in training machine learning models and realize dramatically lower cost compared to AWS EC2 GPU instances.

Resources: Website

AWS Outposts 1U and 2U Servers

What is it?

AWS Outposts is a fully managed service that extends AWS infrastructure, AWS services, APIs, and tools to virtually any datacenter, co-location space, or on-premises facility for a truly consistent hybrid experience. AWS Outposts is ideal for workloads that need low latency access to on-premises applications or systems, local data processing, and to securely store sensitive customer data that needs to remain anywhere there is no AWS region, including inside company-controlled environments or countries.

AWS Outposts 1U and 2U form factors are rack-mountable servers that provide local compute and networking services to edge locations that have limited space or smaller capacity requirements. Outposts servers are ideal for customers with low-latency or local data processing needs for on-premises locations, like retail stores, branch offices, healthcare provider locations, or factory floors.

AWS will deliver Outposts servers directly to you, and you can either have your onsite personnel install them or have them installed by a preferred third-party contractor. After the Outposts servers are connected to your network, AWS will remotely provision compute and storage resources so you can start launching applications.

AWS Outposts 1U and 2U form factors will be available in 2021. To receive more information about Outposts servers, sign up here.

Availability:

At GA, Outposts can be shipped to and installed in the following countries: NA – US; EMEA - All EU countries, Switzerland, and Norway; APAC - Australia, Japan, and South Korea.

Use Cases:

  • Low Latency: Customers with low latency requirements need to make near real time responses to end user applications or have to communicate with other on-premises systems or control on-site equipment. They have adopted the Amazon cloud for centralized operations but need to run compute, graphics, or storage intensive workloads on premises to execute localized workflows with precision and quality.
  • Local Data Processing: Customers that need to access data stores that will remain on-premises for a time. Some customers run data intensive workloads that collect and process hundreds of TBs of data a day. They would like to process this data locally to respond to events in real time and to have better control on analyzing, backing up, and restoring the data.

Key Verticals

  • Manufacturing Automation: Use AWS services to run manufacturing process control systems such as MES and SCADA systems and applications that need to run close to factory floor equipment.
  • Health Care: Apply analytics and machine learning AWS services to health management systems that need to remain on premises due to low latency processing or patient health information (PHI) requirements.
  • Telecommunications: Use cloud services and tools to orchestrate, update, scale, and manage the lifecycle of Virtual Network Functions (VNFs) across cloud, on premises, and edge.
  • Media & Entertainment: Access the latest GPU innovations on premises for graphics processing, audio and video rendering.
  • Financial Services: Build next-generation trading and exchange platforms that serve all participants at low latency.
  • Retail: Leverage AWS database, container, and analytics services to enable retail innovations such as connected store experiences, run point-of-sale systems to process in-person transactions locally.

Customer Benefits:

  • Run AWS Services On – Premises
  • Store and Process Data On – Premises
  • Truly Consistent Hybrid Experience
  • Fully Managed Infrastructure

Resources: Website

AWS Local Zones

What is it?

AWS Local Zones are a type of AWS infrastructure deployment that places AWS compute, storage, database, and other select services closer to large population, industry, and IT centers. With AWS Local Zones, you can easily run latency-sensitive portions of applications local to end-users in a specific geography, delivering single-digit millisecond latency for use cases such as media & entertainment content creation, real-time gaming, live video streaming, AR/VR, and machine learning inference.

Each AWS Local Zone location is an extension of an AWS Region where you can run your latency-sensitive applications using AWS services such as Amazon Elastic Compute Cloud, Amazon Virtual Private Cloud, Amazon Elastic Block Store, Amazon Elastic Container Service, and Amazon Elastic Kubernetes Service in geographic proximity to end-users. AWS Local Zones provide a high-bandwidth, secure connection between local workloads and those running in the AWS Region, allowing you to seamlessly connect back to your other workloads running in AWS and to the full range of in-region services through the same APIs and tool sets.

You can build and deploy applications using AWS services in proximity to your end-users and reduce end-to-end throughput needs for your applications. AWS Local Zones are managed and supported by AWS, bringing you all of the scalability, and security benefits of the cloud. With AWS Local Zones, you can easily build and deploy latency-sensitive applications closer to your end-users using a consistent set of AWS services and pay only for the resources that you use.

Availability:

AWS Local Zones are generally available in Los Angeles, CA and in preview in Boston, Houston, and Miami. Get started with the LA Local zones here. Customers can sign up for access to the preview for Local zones in Boston, Houston, and Miami here.

Use Cases:

  • Media & Entertainment Content Creation: Run latency-sensitive workloads, such as live production, video editing, and graphics-intensive virtual workstations for artists in geographic proximity to AWS Local Zones.
  • Real-time Multiplayer Gaming: Deploy latency-sensitive game servers in AWS Local Zones to run real-time multiplayer game sessions and maintain a reliable gameplay experience. With AWS Local Zones, you can deploy your game servers closer to your players than ever before for a real-time and interactive in- game experience.
  • ML: Easily host and train models continuously for high performance low-latency inference at the edge. Work with your data, experiment with algorithms, and visualize your output faster in AWS Local Zones.
  • Video Streaming: Live stream video content with single digit milli-second latency and high fidelity to your end users. Perform computation and analysis of your video content close to the event and seamlessly extend across Availability Zones and AWS Local Zones close to your end users for high fidelity streaming.
  • AR/VR: Support AR/VR applications by performing computation and analysis close to your end users with AWS Local Zones. Effectively reduce the Motion to Photon (MTP) latencies to the <20 ms benchmark needed to offer a realistic customer experience.

Customer Benefits:

  • Low latency to local end-users: AWS Local Zones place compute, storage, database, and other select AWS services closer to end-users to enable you to open up new possibilities and deliver innovative applications and services that require single-digit millisecond latencies for more end users.
  • Consistent AWS experience: AWS Local Zones enable you to use the same AWS infrastructure, services, APIs, and tool sets that you are familiar with in the cloud. Applications also have fast, secure, and seamless access to the full breadth of services in the parent region.

Resources: Website

Amazon EC2 M5zn Instances

What is it?

Amazon EC2 M5 Instances are the next generation of the Amazon EC2 General Purpose compute instances. M5 instances offer a balance of compute, memory, and networking resources for a broad range of workloads. This includes web and application servers, small and mid-sized databases, cluster computing, gaming servers, caching fleets, and app development environments. Additionally, M5d, M5dn, and M5ad instances have local storage, offering up to 3.6TB of NVMe-based SSDs.

Customer Benefits:

  • Flexibility and choice: Choose between a selection of 60 different instance choices with multiple processor options (Intel Xeon Scalable processor or AMD EPYC processor), storage options (EBS or NVMe SSD), network options (up to 100 Gbps), and instance sizes to optimize both cost and performance for your workload needs.
  • Lower TCO: By leveraging the higher number of cores per processor, M5 instances provide customers with a higher instance density than the previous generation, which results in a reduction in per-instance TCO. With the largest instance size of 24xlarge, customers can scale-up and consolidate their workloads on a fewer number of instances, to help lower their total cost of ownership.
  • Maximize resource efficiency: M5 instances are built on the AWS Nitro System, a combination of dedicated hardware and lightweight hypervisor which delivers practically all of the compute and memory resources of the host hardware to your instances for better overall performance and security.

Resources: Website

Amazon EC2 R5b Instances

What is it?

Amazon EC2 R5 instances are the next generation of memory optimized instances for the Amazon Elastic Compute Cloud. R5 instances are well suited for memory intensive applications such as high-performance databases, distributed web scale in-memory caches, mid-size in-memory databases, real time big data analytics, and other enterprise applications. Additionally, you can choose from a selection of instances that have options for local NVMe storage, EBS optimized storage (up to 60 Gbps), and networking (up to 100 Gbps).

Customer benefits:

  • Flexibility and Choice: Choose from a selection of almost 60 different instance choices with options for processors (Intel Xeon Scalable processor or AMD EPYC processor), instance storage (NVMe SSD), EBS volumes storage (up to 60 Gbps), networking (up to 100 Gbps), and instance sizes to optimize both cost and performance for your workload needs.
  • More memory: R5 instances support the high memory requirements of certain applications to increase performance and reduce latency. R5 instances deliver additional memory per vCPU and the largest size, r5.24xlarge, provides 768 GiB of memory, allowing customers to scale-up and consolidate their workloads on a fewer number of instances.
  • Maximize resource efficiency: R5 instances are powered by the AWS Nitro System, a combination of dedicated hardware and lightweight hypervisor, which delivers practically all of the compute and memory resources of the host hardware to your instances. This frees up additional memory for your workloads which boosts performance and lowers the $/GiB costs.

Resources: Website

 

Recent Posts