VMware vsphere 5.1 Cluster Overview (i) Introduction of vsphere High availability

About the author

Duncan Epping is the chief architect of VMware Technology Marketing, Dun is the architect of vsphere Ha and VMware vcloud Cloud Suite, the first certified design expert of VMware (VCDX 007) ... (There is no more translation here)

Tim2009 that this is a good professional book in VMware Virtualization, so use Google, Youdao, college English combined with virtualization project practice translated into Chinese, hope useful.

Knowledge points

This book is about VMware products, only for personal opinion, the content of the article is not approved by VMware, and there is no need to reflect VMware's view, this is the author's book, not the book of VMware

First of all, we are very grateful to the VMware Management team (Kaushik Banerjee, Jim Senicka, Bogomil Balkansky and Raghu) for their support for this book and related projects.

Objective

Data centers typically require careful configuration, design, monitoring, and other operational tasks to enable efficient running of applications. However, the application of these activities is essentially to ensure that the lights are always bright, there is no significant additional value, not to make IT services into the core objectives of the business, many tasks to manually complete the setup. If the owner of the application defines a high standard of business needs, the IT department translates to the level of the system, interprets these policy requirements with the underlying infrastructure, and automatically configures, manages, and operates the underlying system to match the policy goals, which is the goal of VMware, the new technology solution to build the next generation of data centers.

Data centers are separated into two layers independently-we can consider what to do and how to do it, the upper zone runs their work at the same time matching the policy goals-which explains what to do, and VMware's goal is to make things as automated as possible, and the owner of the application doesn't have to know exactly which physical host the virtual machine is running on, Or what valid rules are applied by the virtual machine, the owner of the application wants the appropriate design and configuration to replace the specified requirements and expectations, so that services can be provided to meet these requirements

The underlying strategies for "How to do", such as cluster high-availability, dynamic resource allocation are defined to this layer, and are programmed to manage. With clustering we no longer need a single server for work, which is the subject of this book and why this is so important to the success of vsphere deployments, VMware vsphere needs to configure the appropriate cluster level to manage resources, and also to achieve higher service levels, The cluster controls a large number of server-forming pools to coordinate resource allocations that may include availability, storage I/O requirements, and other specifications related to the application owner

Therefore, the data center of the new age is not only because of virtualization, it integrates many traditional basic factors such as computing, storage, network components, so that they can be managed by the system, which means that the manual configuration, change, release, are now managed by the system, and some projects and virtualization software architecture are materialized, Modern software provides flexibility and efficiency when job requirements change, and the underlying architecture adapts and expands to meet demand.

Over a long period of time, we expect more and more traditional infrastructures to be managed by the system, and a high level of architecture is like a patch, and application-centric has become part of standard services, preferably at the system level.

The network is becoming more and more representative, for example, the security rules will be assigned to the application, and the contrast is that the unique device is considered a network port group, and similar innovations occur in the storage area, we consider the virtual machine as the unit of data storage, we plan to migrate the associated data from physical memory, such as LUNs or on a file server.

The goal of cloud computing is for all similar infrastructure to continue to be managed in a dynamic manner, with a careful underlying design that requires professional knowledge and provides a book worth reading to help IT administrators focus more on configuring applications to meet business needs, rather than just maintaining the IT infrastructure to keep the machine lights on.

We want you to find a useful and meaningful place in this book.

Raghu Raghuram

Executive Vice President, Cloud infrastructure and Management

Vmware

The first part vsphere high availability

The first chapter introduces vsphere high availability

Availability is one of the most important aspects of traditional data center considerations when delivering services, and VMware has designed a VMware vsphere HA (high-availability) feature to meet the exponential growth of downtime as it delivers services on VMware vsphere shared platforms, The VMware vsphere ha (High availability), hereinafter referred to as HA, provides a simple and cost-effective solution to improve the operating system of highly available and unrelated virtual machines applied on virtual machines. The HA configuration can be completed in a simple few steps of the vcenter server, while providing a unified and simple interface that allows you to turn on the HA function when a esxi or ESX server is set up, and we will use the ESXi server as a reference in this book, That is to say, the ESXi server is the standard. Enabling features can protect your virtual machines, and when a host fails in a cluster, the affected virtual opportunities are automatically restarted on the other ESXi hosts in the same cluster.

Figure 1:ha's Activity flow