Automating your failover, backup and recovery in your virtual infrastructure ensures that you can recover from unplanned hardware, software, or user errors with minimal downtime and less complexity than traditional failover, backup and recovery processes.
Get Automatic Recovery with High Availability (HA) Clusters
Pool your hardware resources from multiple VMware ESX servers and get automatic recovery from system failures with VMware High Availability (HA) clusters. Manage the hardware resources of multiple VMware ESX hosts without requiring dedicated cluster administrators or standby hardware, as required by traditional clusters. Extend the breadth of your application coverage through VMware HA clusters to all your application services, rather than limiting high availability to a few mission-critical services.
Creating VMware HA clusters is much simpler than setting up traditional clusters. VMware vCenter Server, the centralized management core of VMware vSphere, includes a helpful Cluster Wizard that walks you through the steps of creating a VMware HA cluster. Once your cluster is created, you can quickly add and subtract VMware ESX hosts and virtual machines, set priority for individual virtual machines and make restart decisions. When you add a new VMware ESX host, any excess capacity beyond what is required for each virtual machine running on that host VMware ESX is added to your cluster’s resource pool. You can easily see information about your clusters and virtual machines assigned to that cluster, configuration details, failover settings and flags for over committed clusters, making management of your VMware HA clusters much easier when compared to traditional cluster solutions.
Enable LAN-free Backup
There are three basic ways to backup and protect the data in your virtual infrastructure: Agent-based backup, Service Console-based backup and LAN-free backup with VMware Consolidated Backup.
Agent-based Backup
Agent-based backup in virtual machines works like it does in non-virtualized data centers. A backup agent is installed in each virtual machine. A central backup server triggers the agent according to a defined schedule and a copy is made of the virtual machine, then stored on the central backup server. This is resource intensive and requires management overhead to maintain the agent in each virtual machine.
Service Console-based Backup
Having an agent running on the Service Console in VMware ESX is an improvement on agent-based backup because it provides lower management overhead and more efficient backups. Only one agent is required to backup multiple virtual machines. However, it is still a slow streaming backup and is resource intensive. Another drawback is that not all vendors support agents inside the Service Console.
VMware Consolidated Backup (VCB)
If you have shared storage, you can use VMware Consolidated Backup (VCB) to back up virtual machines from a centralized backup server, with no agent needing to be installed on the virtual machines or in the Service Console. VCB takes a snapshot of the virtual machine and mounts it on the VCB backup proxy server. The backup process is then run on the proxy server, so there is very little impact on the virtual machine resources.
Resiliency to Unplanned Component Failures
The VMware ESX hypervisor platform incorporates component-level, failure-handling capabilities by design since the inception of the product. Built-in storage access multipathing ensures shared storage availability with SAN multipathing for Fibre Channel or iSCSI SAN, and NIC teaming for NAS. Moreover, enhanced NIC teaming provides each networked virtual machine with built-in NIC failover and load balancing that enables greater hardware availability and fault tolerance. New NIC teaming policies allow users to configure multiple active and standby adapters. Teaming configuration may be different for different port groups on the same virtual switch and different groups can even select different teaming algorithms for the same team.
Hardware-independent Disaster Recovery
Traditional disaster recovery solutions are generally expensive and unreliable. They require expensive hardware duplication at designated recovery sites, use processes that are manual, complex and slow, and have a high rate of failure due to hardware dependencies and untested failover processes. A virtual infrastructure enables you to transform your disaster recovery plan and ability to failover your services to your recovery site with a high degree of confidence.
With a virtual infrastructure, you can quickly, reliably and cost-effectively recover production applications across the range of your x86 infrastructure rather than limiting your recovery options to only the most critical applications. VMware virtual machines are hardware-independent, greatly simplifying your recovery site needs. With the entire server saved as an image, application, data and configuration information is more readily migrated to new machines, leading to much improved recovery time objectives (RTOs) when compared to failing over services in the physical world. In short, with a virtual infrastructure, your disaster recovery plans can include more systems for less cost, more reliably.
Automating Site-wide Disaster Recovery with VMware vCenter Site Recovery Manager
"IT Service Continuity" is a new term we are using to describe a product called VMware vCenter Site Recovery Manager that delivers pioneering disaster recovery management and automation for the datacenter. VMware vCenter Site Recovery Manager works integrates with VMware vSphere and VMware vCenter Server to make recovery rapid, reliable, affordable and manageable by making it possible for customers to build, test, and automate disaster recovery plans.
Take the Next Step
Discover how to implement robust failover and load balancing or learn more about using VMware HA and VMware DRS for ensuring the availability of critical IT services and applications.
