This special edition includes the most important announcements and major updates regarding Azure infrastructure as a service (IaaS) and Azure Stack, officialized by Microsoft this week during Microsoft Ignite 2020 conference. Microsoft announced several important additions to its Azure infrastructure as a service (IaaS) portfolio and the Microsoft infrastructure services continue to evolve to optimize the experience of running business-critical workloads.
Availability Zones deployed to more Azure Regions
Azure Availability Zones (AZs) provide a high-availability option for comprehensive business-continuity and disaster-recovery strategies (BCDR), 99.99% uptime service-level agreement (SLA) on virtual machines, flexible high-performance architecture and multizone support with built-in security. Microsoft continues to expand deployment of Availability Zones in datacenter regions worldwide, with a roadmap to provide Availability Zone options in each country it operates datacenters in over the next 24 months. In September, Availability Zones will be available in two more existing regions, Canada Central and Australia East, bringing the total number of Microsoft’s AZ-enabled regions to 14.
Zone to zone disaster recovery for azure virtual machines
Zone to Zone disaster recovery allows customers to replicate, failover and failback their business-critical virtual machines within the same region with zones. The capability adds options for scenarios such as metro-based disaster-recovery strategy while customers are hosting applications on-premises and are looking to mimic that after migrating applications over to Azure; those that have complicated networking infrastructure and want to avoid the cost and complexity of recreating it in a secondary region; and those in regions that prefer not to use paired region disaster recovery options.
New Azure Spot VM features
New Azure Spot VM features, in preview soon in the Azure portal, enable customers to access and review the price history of Spot VMs and eviction rate for the past 28 trailing days. Also, not only allow customers to derive estimates about the probability that their workloads will be evicted, but also enable better estimates for the cost of running interruptible workloads using Spot VMs.
Azure Hybrid Benefit for Linux workload (preview)
Azure Hybrid Benefit, available in preview, improves flexibility and enhances user experience for Red Hat and SUSE customers migrating Linux to Azure.Directly in the portal or through CLI, Red Hat Enterprise Linux (RHEL) and SUSE Linux Enterprise Server (SLES) customers will be able to convert existing Linux VMs from pay-as-you-go (PAYG) billing to bring-your-own-subscription (BYOS) billing, making use of their existing Red Hat and SUSE subscriptions. This is a unique capability that allows customers to first deploy a POC in Azure using the convenience of on-demand PAYG Linux VMs, and when testing is complete, convert it to long-term production using RHEL and SLES subscriptions. This removes the headache of production redeployment, preserves existing investments in on-premises RHEL and SLES subscriptions, and reduces migration planning worries.
A new Azure-supported Linux distribution
Flatcar Container Linux by Kinvolk, is now available in Azure Marketplace. Flatcar is an immutable Linux distribution and is compatible with Core OS (which reached its end of service on May 26, 2020), making Flatcar Container Linux a viable and straightforward migration choice for container workloads running on Azure.
Azure Image Builder
Azure Image Builder, generally available by the end of this year, is a free image-building service that streamlines the creation, update, patch, management and operation of Linux and Windows images. Azure Image Builder will deploy resources into your subscription when used, and you pay only for the virtual machines and associated storage and networking resources consumed when running your image-building pipeline.
Multiple new Azure Infrastructure features
Multiple new Azure Infrastructure features are now available:
- New Azure Virtual Machines (VMs) are now generally available featuring Intel Cascade Lake processors for general purpose and memory-intensive workloads. These VMs provide up to 20% greater CPU performance compared to the prior generation.
- Azure Dedicated Host now gives customers more control. Customers can schedule host maintenance operations on Dedicated Hosts and isolated VMs as well as control when guest OS image updates are rolled out. Azure Dedicated Host also supports Virtual Machine Scale Sets and simplifies deployment by offering customers the ability to let the platform select the host group where VMs are deployed to.
New Azure Disk Storage updates
New Azure Disk Storage updates, including:
- Azure Private Link integration which enables secure import and export of data over a private virtual network for enhanced security
- Support for 512E on Azure Ultra Disks to enable migration of legacy databases to Azure.
Cisco SD-WAN with Azure Virtual and Global Load Balancer feature (preview)
Azure networking enhancements announced at Ignite include the addition of Cisco Software-Defined Wide Area Network (SD-WAN) native support within the Azure Virtual WAN hubs, and the Global Load Balancer feature for Azure Load Balancer. Both are available in preview.
The use of Cisco SD-WAN with Azure Virtual WAN aligns with networking trends to leverage technologies such as SD-WAN to improve performance through intelligent path selection and central policies. They work to eliminate traditional networking backhauls by sending traffic directly from branch to the cloud via local breakouts and allow you to leverage your chosen vendor’s path selection and policy management.
With Global Load Balancer, customers can use the feature in the Azure Load Balancer to distribute traffic to their global applications, improving performance and availability.
Azure orbital: a new managed service that provides access to physical satellite communication (private preview)
Azure Orbital is a new managed service that provides access to physical satellite communication capabilities to process and analyze data in Microsoft Azure. Take advantage of a low-latency global fiber network when working with large satellite datasets. Azure Orbital is available now to select customers in private preview. Azure Orbital enables satellite operators to schedule contacts with their spacecrafts and directly downlink data into their virtual network (VNet) in Azure.
Azure Stack Edge
Two new Azure Stack Edge rugged devices are available
Customers can perform machine learning and gain quick insights at the edge by running the Azure Stack Edge Pro R with NVIDIA’s powerful T4 GPU and the lightweight, portable Azure Stack Edge Mini R. Both devices are designed to operate in the harshest environments at remote locations.
Azure Stack Edge is now available with GPUs
Customers can run visualization, inferencing, and machine learning at the edge with the Azure Stack Edge Pro series powered by the NVIDIA T4 Tensor Core GPU. This unlocks a broad set of new edge scenarios, such as automatically recognizing license plates for efficient retail curbside pickup, and detecting defects in real time in products on a manufacturing assembly line.
Azure Stack HCI
Preview of Azure Kubernetes Services (AKS) on Azure Stack HCI
AKS on Azure Stack HCI enables customers to deploy and manage containerized apps at scale on Azure Stack HCI, just as they can run AKS within Azure. This now provides a consistent, secure, and fully managed Kubernetes experience for customers who want to use Azure Stack HCI within their datacenters. Sign up for the preview of AKS on Azure Stack HCI.
Azure Stack Hub
Azure Stack Hub is now available with GPUs
To power visualization intense apps, we’ve partnered with AMD to bring the AMD Mi25 GPU to Azure Stack Hub, which allows users to share the GPU in an efficient way. The NVIDIA V100 Tensor Core GPU enables customers to run compute intense machine learning workloads in disconnected or partially connected scenarios. The NVIDIA T4 Tensor Core GPU provides visualization, inferencing, and machine learning for less compute intense workloads