Tuesday, January 15, 2019

Industry-Specific Database Platforms Terms

A server is a physical or virtual machine hosting SQL Server software complete with an operating system.

A cluster is a set of servers grouped together to provide redundancy. The cluster manager monitors the health and responsiveness of the cluster members (nodes). On Windows Server, the manager is called the WSFC (Windows Server Failover Cluster [manager]). Linux distributions use Pacemaker as a cluster manager. A cluster is managed at the operating system level.

A node is a member of a failover cluster. Each node in the cluster has the same capability to answer requests and has access to the same data as every other node in the cluster. Each node must be available to answer requests in a timely manner to be considered “healthy” by the cluster manager.

A database is a data structure that stores application data and metadata in physical files on disk.

An instance is a collection of SQL Server databases, jobs, and so forth run by a single SQL Server service that runs in memory on a specific computer at a specific time. An instance is accessed using a single IP address and all requests are sent to that IP. The source of the request doesn’t know which physical location the response it receives is coming from.

A replica (or database replica) is a set of two or more SQL Server instances that are grouped together in a failover configuration. In other words, a replica is—at the database level—what a cluster is at the operating system level.

Popular Posts

Get Sponsored by Big Brands