Microsoft cluster solution (I)

Microsoft has been expanding the scalability, availability, and reliability of its server solutions for several years. Cluster technology has proved to be an effective way to achieve this goal. Microsoft fully supports the cluster technology concept and is committed to making it an integral part of the integration of Microsoft operating systems and related products. With the official release of Windows 2000, Microsoft's cluster solution has gradually entered a mature stage.
Scalability, availability and reliability
A cluster is a group of independent computers that run the same set of applications in collaboration and provide a single system image for clients and applications. The purpose of cluster technology is to further improve the scalability, availability and reliability through multi-layer network structure.
Scalability refers to the ability of a computer to process increasing workloads while maintaining acceptability. Hardware Device scalability implicit scaling, according to Microsoft) relies on large computers with scalability to perform various operations. Software Product scalability extended) depends on the Cluster Formed by multiple computers that are organized together through collaborative work. It is different from RAID drive arrays. In fact, Microsoft uses an informal term computer redundancy array RAC) to name its own extended expansion cluster. Just as you can improve performance by adding disks to a RAID array, you can also improve performance by adding nodes to an extended group.
Availability and reliability are two closely related but slightly different concepts. Availability refers to the existence of quality, backup capability, accessibility and accessibility. Reliability refers to the system strength. Even the most reliable system always has problems one day. Hardware equipment manufacturers take advance preparations for possible failures by providing redundancy in key technical areas such as disk drives, power supply devices, network controllers, and cooling fans. However, the redundancy provided on a computer cannot prevent application faults for users. If the database software on a server fails, although the server may be very reliable, functions provided by combining the software with the server will still be unavailable. Therefore, a single computer cannot meet all the necessary scalability, availability, and reliability challenges undertaken by the cluster.
The cluster can simulate RAID arrays to provide availability and reliability. In a fault-tolerant disk configuration scheme such as RAID 1 or RAID 5, all disks work collaboratively in Redundant Arrays. If a disk fails, you can unplug it and insert it into a new disk, and the rest of the disk will continue to run-No configuration, no installation required, and most importantly, will not cause downtime. The RAID system automatically reconstructs the new drive so that it can work with other drives. Similarly, when a computer in the cluster fails, you only need to replace it with a new system, and the entire cluster continues to run. Some Cluster software can automatically configure the server and integrate it into the cluster-all related operations are completed when the cluster is available.
Four cluster Solutions
Microsoft provides four basic cluster technologies: Microsoft cluster service (MSC), network load balancing (NLB), Component Load Balancing (CLB), and Application Center 2000. These services are provided through three solutions: MSC, NLB, and Application Center. CLB is an integral part of Application Center and can only be applied through Application Center. NLB can be used either through Application Center or as an independent solution. Windows 2000 Advanced Server and Windows 2000 Datacenter Server contain both MSC and NLB, but you must purchase Application Center separately.
Table 1 summarizes the availability of these four cluster technologies in different Windows 2000 Server and Windows NT Server 4.0 product families. As you might imagine, none of these technologies apply to Windows 2000 Professional or Windows NT Workstation 4.0.
　
　
Microsoft cluster service
Initially codenamed Wolfpack and known as Microsoft Cluster Server and Microsoft cluster service, the MSC is Microsoft's first major attack in the NT cluster technology field. It is recognized as the best Microsoft cluster solution. In the MSC cluster, the MSC Software can establish connections with up to four physical computers running on the high-speed network. Generally, computers in a cluster can share the same storage subsystem and functions in the "Active-active" mode, which means all cluster computer nodes) you can actively complete the work by sharing the load, and share the work of a node in case of a fault. Figure 1 shows a 4-node MSC cluster.

Figure 1 4-node cluster implemented by Windows 2000 MSC
The main purpose of MSC is to improve application availability through its fault tolerance capabilities. Fault Tolerance refers to the function of porting a faulty application on a node to a cluster on another healthy node in the cluster due to hardware device failure or software error. After the faulty application is restored, the cluster should be able to implement "fault return" for the original cluster node ". On the premise that no data related to the faulty application is lost, the MSC can recover the applications running on the cluster and manage the fault return, it also maintains the user and application status during the fault recovery process. This type of cluster function is called the stateful cluster function. In contrast, NLB, CLB, and Application Center provide stateless clustering and Dynamic Load Balancing while enhancing availability. I will discuss this in detail later ).
For applications such as email servers and database applications, MSC is a good way to run. Suppose you decide to run Microsoft Exchange 2000 Server on a 4-node MSC cluster. After you install the MSC Software and the Exchange 2000 version for the cluster, you can configure the cluster so that Exchange 2000 can perform fault recovery on the backup node when the primary node fails. When a fault occurs, there must be a user session in the active state on the master server. However, MSC can quickly and automatically restore the fault without losing any data. The backup node will replace the workload and related data from the faulty node and continue to provide services to users.
At the same time, you can continue to work during application upgrade. You can use a rolling upgrade method, such as upgrading applications on one cluster node each time and ensuring that applications on other nodes are still available. You do not have to stop using applications during the upgrade process. For example, assume that you have a dual-node cluster. Node 1 runs Exchange 2000, and node 2 runs Microsoft SQL Server. You want to configure this cluster so that Exchange 2000 and SQL Server can implement fault recovery between each other if necessary. When you need to upgrade SQL Server, you can start the SQL Server fault recovery function on node 2 through the MSC Cluster Manager. When Node 1 takes over the SQL Server running task and continues to run Exchange 2000), you can upgrade SQL Server Software on node 2. After the upgrade is completed, you can migrate SQL Server from node 1 to node 2 by returning the fault, the same software upgrade operation is performed repeatedly on SQL Server on node 1. After Node 1 completes the upgrade, the entire SQL Server Software completes the upgrade task without affecting user usage.
Unlike the other three Microsoft cluster solutions, you generally cannot use MSC to expand applications for more users. The MSC cluster cannot provide dynamic load balancing capabilities between nodes or achieve distributed Application running to NLB, CLB, and Application Center in the stateless non-shared mode. In fact, the only feasible way to scale applications by using MSC is to manually allocate applications to different cluster resources during the installation process. For example, to serve 2000 users on the Exchange 5000 platform, you can apply a 2-node activity-active cluster and provide services for 2500 users on each node. In this way, you can not only obtain the performance advantages of providing services to users through two servers, but also realize the necessary availability in case of faults. However, when a fault is restored, the remaining node must be able to provide services to all 5000 users before the faulty node is resumed.