Data Center Cloud Computing Tour

Today, if you don't know about cloud computing, someone will say you're out, because all of it is talking about cloud computing. But how does cloud computing connect with existing data centers? How do you get the Express on cloud computing from existing data centers? I do not want to explore the concept of cloud computing here, and from the data Center cloud computing trip to analyze the impact of cloud computing.

Before exploring the data center's cloud-computing journey, it is necessary to clarify a myth: Many people think that Amazon, Microsoft and so on have built the cloud platform, directly to the existing data center to move up the cloud is not started? We all know that cloud computing is divided into private cloud and public cloud, Google, Microsoft and so on are built by the public cloud, at the same time, cloud computing can be divided into proprietary cloud and common Cloud, Google, Microsoft and so on build is proprietary cloud. We can easily migrate existing applications to the common public cloud, but the proprietary public cloud cannot afford it. Proprietary cloud needs to be developed specifically to meet applications developed based on specific conditions. Traditional applications to migrate to proprietary cloud, often need to modify the source code. And we have a lot of applications in our existing data centers that don't have the source code, even if you have the source code, you may not be able to find the development team of the year to modify it, which makes the proprietary cloud only suitable for certain application scenarios, for our existing data center dozens of, or even hundreds of applications, often powerless. All, the data center of the cloud trip can not take the exclusive cloud this train. So where should the data center's cloud trip start?

First, the data center's cloud-computing journey starts with building a resource pool. Without the pooling of resources, cloud computing would be impossible to talk about. The so-called pooling of resources, including computing resources, network resources, storage resources of the pool. Only after the pooling of resources can flexible configuration, dynamic adjustment, to be able to have flexibility, improve stability and security. Resource pooling is the first and particularly important step in the entire cloud. When choosing a resource pooling scenario, consider the following five areas:

Built on a reliable, proven basis

Provides a platform for shared IT services

Complete solutions for virtualization and cloud management

Support your entire IT infrastructure

Validated in tens of thousands of customer deployments

VMware's Vsphere 4.0, released last year, is the industry's first cloud operating system and the best tool for resource pooling. Vsphere 4.0 has already been able to provide some cloud computing capabilities, and Vsphere 4.1, released in July this year, has been further upgraded from all levels, with the introduction of memory compression technology, storage and network I/O control capabilities to further improve efficiency, control, and selectivity, More to meet the needs of cloud computing platforms. VMware vsphere 4.1 has been fully enhanced from the above five areas to fully meet the needs of the data center to build a public cloud for the private cloud and cloud service providers, and recently VMware, VMworld 2010 in San Francisco, has released and Bluelock, Colt, SingTel, Terremark and Verizon provide VMware Vcloudtm Datacenter Service, an important success story for VMware Cloud Solutions, marking the VMware vsphere and VMware VCLOUDTM director become the most trusted base platform for both private and public clouds. The new features of Vsphere 4.1 are shown in the overview diagram below.

The second step in the data center cloud computing journey is policy configuration and automation. With cloud computing, you don't have to spend valuable time managing your physical infrastructure or applications to focus on innovation and deliver value to your business. Once all hardware resources are concentrated in the virtual building blocks and the service levels of the associated application are defined, the automation plan can be deployed based on business rules and application requirements. Automation improves IT's ultimate efficiency by mapping plans and policies to the business and defining mappings in the private cloud's service delivery mechanism (applications and services delivered according to service level). The process of automating the data center's cloud computing journey can be summarized as follows:

· Define services, including complete virtual application stacks (Vapp and single servers (applications, Web servers, database servers) and media (operating systems, software, databases, application servers, etc.)

· Set service levels: Availability (uptime) assurance, resource access, storage speed, network bandwidth, computing resource share, storage IO assurance, backup frequency, and disaster recovery requirements;

· Set rules for automating load balancing and optimizing resources, enabling applications to benefit from dynamic resource scheduling, uptime guarantees, and disaster recovery failover rules;

· Tiered service levels based on application and business priorities, and develop policies for self-service access to applications and resources, allowing users to manage seamlessly in a private cloud;

· Create categories of these services to allow self-service access and generalization of policies and permissions (user rights) for accessing, creating, and modifying these services, including pre-built, direct deployment of virtual machines, vApp, media, and software.

· Determining the price and billing methods of these services;

· You can automate administrative tasks by scripting with APIs such as the Vsphere Management API (VIM) and the Vcloud Developer API.

By applying these automation rules to the cloud infrastructure, you can efficiently implement a zero-touch infrastructure where IT can serve as a service provider (ITaaS). IT will be able to better manage risk by using cloud computing methods that keep the security, control, and compliance required for the enterprise.

(Responsible editor: The good of the Legacy)