Veritas Cluster Server

from Wikipedia, the free encyclopedia

Veritas Cluster Server (VCS) refers to a software product from Veritas Software Corporation for the interconnection of several computers to form a computer network (cluster). In 2016 the cluster became part of a new product bundle and is now called Veritas InfoScale Availability.

functionality

This software connects two or more computers, regardless of the operating system, to a so-called computer cluster . Computers are also commonly referred to as nodes in a cluster system. The computers that are interconnected to form a computer network usually increase the reliability of the overall system for general tasks as a server due to the cluster formation.

One of the special features of the Veritas cluster is that dependencies between service groups can be modeled quite easily (via GUI). Using phantom resources, dependencies between resources in different service groups can also be formulated.

Another special feature is the independence from a specific operating system. The Veritas cluster software can run on

can be used with the "Veritas Storage Foundation HA". The prerequisite, however, is that the same operating system (one from the selection mentioned) has been installed on all cluster nodes involved.

The database versions 9i, 10g, 11 and 12 from Oracle are approved for the cluster software version for databases "Storage Foundation for RAC". The existence of two heartbeat connections between the nodes involved is required. The use of other database systems (such as DB2 / MySQL or similar) is also possible in principle.

The software product runs independently of the operating system kernel (also called user-level cluster software) in normal system processes. It can be used in different configurations ( LAN / MAN / WAN ) and i. d. Usually supplemented by application-specific agents. There is also a special version that enables the implementation of Oracle databases in a Veritas cluster ("Veritas Storage Foundation for Oracle RAC ").

Comparable functionality of a “high availability” (HA) cluster system can be found in IBM Tivoli System Automation for Multiplatforms (IBM SA MP), IBM HACMP , Sun Cluster , Linux-HA or Microsoft Cluster Server (MSCS).

Example of a Sun Solaris system with Veritas cluster

The operating system (e.g. Sun Solaris , Windows Server 2003 ) is installed on two host systems (i.e. computers or nodes), which should largely have the same hardware. For computers with the Windows operating system, the version "Veritas Storage Foundation HA" is then installed and configured as a combination product of Veritas Cluster Server and Veritas Storage Foundation .

The hardware should largely match in order to be able to provide the users with sufficient system performance in the event of a system failure (so-called failover ). In order to increase the reliability, it must be ensured that at least two independent network cards are used per cluster node when selecting the hardware . The double (or multiple) continuous availability (eg by. Another critical system components such as storage NAS , SAN ), network components ( switches , wiring) or power (several hot-swap - power supplies , UPS ) increases the availability of the system (see Single point of failure ).

The reason for the two independent network cards is the functionality of each cluster system , the heartbeat . “Heartbeat” means the internal communication of the systems that are supposed to act as cluster nodes. In this way, the two computers inform each other whether the other partner (host or sub-cluster system) is still functional. If this is no longer the case, the other host or sub-cluster system takes over the work.

The function of the software is divided into so-called "service groups", which in most cases provide databases , network file systems ( NFS ) or critical applications for the user.

See also

literature

  • Volker Herminghaus, Albrecht Scriba: Veritas Storage Foundation, Springer (2006), ISBN 3540346104
  • Heiko Bauke, Stephan Mertens: Cluster Computing. Springer 2006, ISBN 3-540-42299-4
  • Greg Pfister: In Search of Clusters , Prentice Hall, ISBN 0-13-899709-8 (Eng.)
  • Evan Marcus, Hal Stern: Blueprints for High Availability: Designing Resilient Distributed Systems , John Wiley & Sons, ISBN 0-471-35601-8 (English)

Web links

Individual evidence

  1. ↑ About global clusters. Retrieved April 26, 2017 .