Service Group Management in Virtual Business Services

Service group management improves business resiliency by providing a method to bundle hardware, software, applications, databases and networks into a single entity with dependencies. By monitoring the health and performance of these service groups, through proactive notification, pending issues can be quickly addressed. VOM reports on the relationship of applications to virtual machines, physical servers and clusters and provides coordinated failover of services that span virtual machines and physical machines for multi-tier applications. In the past, customers who wanted this functionality had to build scripts to automate these procedures but this method was complex to manage and test.

To help customers address these issues, Veritas introduced Virtual Business Services (VBS). Virtual Business Services combines the power of VCS, AppHA and VOM to provide complete multi-tier business service management and High Availability. VBS now enables management of multi-tier business services on top of VOM and VCS which allows VOM to be used as a single tool for availability management.

Virtual Business Services achieves the following:

High Availability is primarily managed within each tier. The cluster is responsible to keep services highly available within the cluster. The boundaries for an application are the cluste instance. Logically, a VBS can be seen as a container that allows service groups to be built into a single object. To enable VBS, Red Hat Enterprise Virtualization Environments must have VCS installed on the physical server. For the other operating environments within the VBS, it is required that each tier has VCS, Microsoft Cluster Server installed.

In order to deploy VBS, there must be at least one VOM Central Server installed in the data center. The VOM Central Server is used for configuration, visualization and management of VBS. However, after the initial configuration of a VBS, it can be managed using a CLI as well. VBS functionality does not depend on VOM Central Server. CLI operations work regardless of whether the VOM Central Server is available or not, and the member nodes of a VBS will operate autonomously of the VOM Central Server once VBS is deployed.

Application DR can be between VMs or from a Virtual to Physical DR and vice versa. During the failover of Virtual Machine there is an automatic update of VM (IP, DNS, netmask) to ensure user access to the new instance.

An example of how DR operates across a multi-tier environment

Figure: DR in a multi-tier environment

DR in a multi-tier environment

Veritas Operations Manager also includes the ability to associate different Virtual Business Services into a Disaster Recovery Plan. This feature enables another level of automation because it allows the customer to combine service groups, Virtual Business Groups and manual scripts into a single procedure. It provides the sequence of operations that will be performed at the DR site, in the event of a disaster. The GUI allows you to choose items to include into the plan and provides single click failover of an entire data center to a secondary site.