云计算与虚拟化技术 第8讲: High Availability 阮晓龙 13938213680/n@hactcm edu cn http://cloud.xg.hactcm.edu.cn http://www.51xueweb.cn 河南中医药大学信息管理与信息系统教研室 信息技术学院网络与信思系统科研工作室 20194
2 讨论提纲 a Introducing High Availability Understanding the Layers of High Availability ■ Clustering vMs a Implementing vSphere High Availability Understanding vSphere High Availability Clusters Understanding VSphere High Availability's Core Components Enabling vSphere HA Configuring vSphere High Availability Managing vSphere High Availability a Planning for Business Continuity 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
3 1. Introducing 1.uNderstanding the Layers of High Availability o Even in non-virtualized environments, there are multiple ways to achieve high availability for Operating System(OS)instances and applications. When you introduce virtualization into the mix with VSphere, you gain additional methods of providing high availability At each layer are tools and techniques for providing high availability and business continuity o At the Application layer, options include Oracle Real Application Clusters RAC), or simply application resiliency through a micro-services architecture o At the OS layer, solutions include OS clustering functionality, such as Windows Server Failover Clustering(WSFC) 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
4 1. Introducing 1.uNderstanding the Layers of High Availability o Even in non-virtualized environments, there are multiple ways to achieve high availability for Operating System(OS)instances and applications. When you introduce virtualization into the mix with VSphere, you gain additional methods of providing high availability At each layer are tools and techniques for providing high availability and business continuity o The virtualization layer offers a number of features for high availability, including vSphere High Availability(HA), vSphere Fault Tolerance(FT), and stretched(metro) storage clusters At the Physical layer, high availability is achieved through redundant hardware--multiple network interface cards(NICs) or host bus adapters (HBAs), multiple storage area network(SAN) switches and fabrics, multiple paths to storage, multiple controllers in storage arrays, redundant power supplies, and so forth 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
5 1. Introducing 1.uNderstanding the Layers of High Availability FIGURE 7.1 Each layer has its own App App App App Application layer forms of high availability H|&: perating Virtualization software Virtualization layer 一一一一一一一一一一一一一一一一一一一一 Physical infrastructure(switches, NICs servers,SANs, disks, etc. Physical layer 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
6 1. Introducing 1.uNderstanding the Layers of High Availability a Each of these technologies or techniques has its own strengths and weaknesses. For example providing redundancy at the Physical layer is great, but it doesn 't help with failures at the Application layer m Conversely, protecting against application failures wont help much if the underlying hardware isn 't redundant a As you set forth to establish high availability for your virtualized workloads, keep in mind that there is no "one size fits all "solution Use the right tool for the job based on your specific requirements 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
1. Introducing 1.uNderstanding the Layers of High Availability o On VSphere, four key technologies or techniques that help provide high availability OS clustering in Microsoft Windows m ESXi host clustering using v Sphere hA a Virtual machine(VM)mirroring using vSphere FT VSAN stretched storage clustering 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
8 1. Introducing 1.2Clustering VA a Because Windows Server is widely used in corporate and enterprise datacenters today, it's quite likely that you've been asked to create or support a Windows-based cluster a There are two primary ways to use clustering to provide hig availability for Windows Server Network Load Balancing(NLB)clustering Windows Server Failover Clustering(WSFC) 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
9 1. Introducing 1.2Clustering VA FIGURE 7. 2 An nlb cluster can contain up to 32 active nodes(only 5 are shown here), and traffic is 田田田田田 distributed equally across each available VM1 VM2 VM3 VM4 VMs node. The nlb software allows the nodes to share a Loadbalancedidentitywww.rainpole.local(10.1.1.10) common name and Ip address that is reference ed by clients 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn
0 1. Introducing 1.2Clustering VA FIGURE 7.3 Public network Server clusters are best Shared identity: sql. rainpole local (10.1.1.20) suited for applica tions and services like SQL Server, DHCP and so on, which use a common dataset Private(heartbeat) network Shared 河南中医药大学/阮晓龙/13938213680/htp/cloudxg.hactcm.edu.cn/http:/www5ucwebcn