Questions? Feedback? powered by Olark live chat software
Ignorar Navegação

Expanding our Hybrid Cloud and Furthering Cloud Hyper-Scale

Publicado em 9 julho, 2015

General Manager, Cloud Platform Marketing
At Microsoft, we are committed to helping you achieve more in the cloud-first world. Across our proven enterprise capabilities, hyper-scale cloud infrastructure, and comprehensive hybrid solutions, we are bringing the benefits of cloud speed, scale and economics to everyone-- no matter what platform you are on, or what cloud environment you choose. Unlike other vendors in cloud computing, Microsoft offers a comprehensive set of offerings spanning Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), on-premises datacenters, and hybrid environments with support for both Windows and Linux. Today we are releasing several new announcements across Microsoft Azure that further our vision of delivering the most complete, open cloud in the industry.

Extending Hybrid Cloud with New Management and Availability Capabilities

We continue to hear from many of you that hybrid cloud is an ideal solution for your organizations. It combines the agility you need to innovate quickly with the control of on-premises datacenters, while allowing you to leverage and extend existing infrastructure investments. To empower you to take advantage of hybrid cloud, today we are introducing the general availability of new heterogeneous workload support for Azure Site Recovery to make recovery, cloud migration and management of multiple environments, including Amazon Web Services and VMware, easier than ever. Built on technology we acquired from InMage just one year ago, this new automated functionality includes:
  • AWS to Azure migration. With AWS, you are often constrained in taking advantage of a consistent and complete hybrid cloud solution. These new capabilities provide an easy onboarding to Azure, along with the hybrid flexibility and freedom of the Microsoft Cloud Platform.
  • Azure Site Recovery provides support for recovery in Azure for VMWare and physical servers, to bolster existing support for Hyper-V workloads. This now enables you to use a secondary site, private cloud or Azure as your disaster recovery site within one Microsoft solution. Extending and protecting VMware workloads into Azure is an easy, automated process and works seamlessly with your existing infrastructure.
Azure Site Recovery, a key component of our Availability on Demand offering will also be available to Microsoft Operations Management Suite (OMS) customers to deliver hybrid cloud management capabilities unmatched by any other cloud vendor. OMS extends the value of System Center and helps you manage and protect your corporate workloads no matter where they run – Azure or AWS, Windows Server or Linux, Hyper-V or VMware. We also recently announced an add-on for System Center customers to more easily take advantage of OMS at one low cost. As the leading hybrid cloud vendor, we’re committed to providing both management technologies and infrastructure to power hybrid clouds around the world. This commitment is evident through other recent announcements at Ignite. For example, with Microsoft Azure Stack, we will bring Azure services in your datacenter to improve service delivery models and increase agility. In addition, Windows Server 2016 is more cloud-inspired than ever with innovations like Nano Server, a purpose-built OS, to power born-in-the-cloud applications, containers, and on-premises software-defined datacenter solutions. We also continue to see growth of hybrid services like ExpressRoute, which has nearly 20 partners and over 100 connectivity providers across the globe. You can expect to hear more from us across both hybrid cloud infrastructure and management, so stay tuned for more updates over the next few months. In the meantime, you can get started with Azure Site Recovery and the other components of OMS today.

Furthering Cloud Hyper-Scale while Enabling Customer Choice

With access to tens of thousands of cores in a single click, the cloud makes hyper-scale computing possible. Today we are further demonstrating how Azure is delivering hyper-scale to the masses across the high performance computing (HPC) space. Customers who prefer the performance and control that is needed for running compute-intensive workloads such as running large scale simulations can rely on Azure Virtual Machines. They can analyze and simulate complex high-fidelity models without having to build, manage, or maintain capital-intensive infrastructure. Today, we are adding support for Linux Remote Direct Memory Access (RDMA) technology with Intel Message Passing Interface (MPI), running on Virtual Machines. With this update, you can now use Windows or Linux to reap the benefits of RDMA, expanding our commitment to openness and choice. We are the only cloud vendor to offer RDMA, which enables virtual machines to communicate over the low latency, high throughput InfiniBand network. This network provides less than 3 microsecond latencies and more than 3.5 Gbps bandwidth. Over the past few months, we have worked closely with Livermore Software Technology Corporation (LSTC) and D3VIEW, to test the performance of the new Linux RDMA technology. These tests have shown near-linear scaling up to 256 cores for an automotive crash. I am also excited to announce that Azure Batch has reached General Availability. Batch is a service that combines our previous GreenButton acquisition with the power of Azure infrastructure, providing the ability to execute large-scale, high-performance computing (HPC) scenarios in Azure. Many of our customers who prefer to focus on getting results faster and not deal with managing or maintaining compute clusters are well-suited to use Batch. Since our public preview, we have seen many customers choose Batch for a range of complex scenarios, from video encoding to risk modeling solutions – all of which have historically been challenges at cloud scale. Towers Watson uses Batch for their Risk Agility FM solution, which is offered to insurers on a subscription basis. Batch has helped reduce the cost and complexity of running massive computational models in Azure, resulting in expected savings of up to 40 percent. In addition, TVEverywhere, a customer who provides video versioning and management services, uses Batch to handle large throughputs needed for huge media files. Batch has allowed each of them to dynamically scale resources, optimizing their systems for improved costs, speed, and quality. Lastly, we are releasing HPC Pack 2012 R2 Update 2 to enable support for Linux Virtual Machines in Azure. HPC Pack allows you to build and manage high-performance clusters in the way that best suits your needs: on-premises, hybrid, or completely in the cloud. If you have already made huge investments in optimizing your on-premises HPC environments, you can dynamically extend them to Azure when additional capacity is needed. Or, you can deploy entirely in Azure and use the wide range of compute- and memory-intensive Virtual Machines without compromising performance. Whether providing you the tools to solve your business problems at cloud scale or helping you maximize your IT investments through our hybrid solutions, we are committed to providing you with the most complete, open cloud to transform your business. Be sure to check out the new features and we look forward to your feedback.