Overcoming the risks associated with rapid adoption of virtualization

By SMBWorld Asia Editors | Mar 23, 2011

0 comments

 

 

 

Digg

 Email

Print

Thumbnail: 

Acronis is warning small and medium businesses (SMBs) about the potential concerns associated with the rapid adoption of virtualization, as the complexity of managing migration, backup and recovery between physical, virtual and cloud environments set in.

The warning stems from results from the recently released Acronis Global Disaster Recovery Index 2011 showing that 73% of SMBs worldwide agree that virtualization has either completely or partially changed the way the business manages its backup and disaster recovery. While the introduction of virtualisation was fuelled by server consolidation and cost efficiency, this so-called next phase or second generation of virtualization adoption poses challenges to traditional backup and recovery processes as users struggle to implement known backup and disaster recovery practices in a new hybrid environment.

“The introduction of server and workstation virtualization was not about backup, it was largely driven by cost and consolidation. As we progress into widespread virtualization adoption, IT managers are learning that traditional physical server backup solutions are inadequate for virtual machine backup, while maintaining separate backup strategies for physical and virtual confuses the backup scenario even more,” explains Bill Taylor-Mountford, President of Acronis APAC.

"Many traditional backups are agent-based, which means that an application is required and consumes precious virtual machine processing resources. Simultaneous initiation of agent-based backups can cause serious virtual machine disruptions, including total failure of the underlying physical host. The next phase of virtualization has to include backup best practices leveraged for a hybrid environment, a central solution for all environments.”

Agent-based software has been adapted to provide some of the functionality required for backup and recovery in a virtualized environment. However, experts warn that these workarounds have proven to be hard to implement, ineffective and added costs to the virtualization programs. These issues combined with the complexity of managing data across physical, virtual and cloud environments are being noted as potential obstacles in the path to effective virtualisation and creating disaster recovery strategies.

To help realise the full benefits of virtualization, Acronis suggests following tips to IT managers:
1.    To close the loop on virtualization efforts, a backup and disaster recovery strategy as robust as that deployed for physical servers is required for all virtual machines, especially if the virtual machine supports a production application.

2.    To provide the lowest virtual machine Recovery Point Objective (RPO) and the fastest Recovery Time Objective (RTO), each virtual machine should be independently backed up and frequently refreshed. This is to ensure time difference between last backup and the current production server state is as small as possible.

3.    Image-based backups provide quick recovery in a cloud environment. Image-based recovery restores the entire virtual machine (VM), including the guest operating system and configuration settings. File or block-based backups only restore data and require the entire VM to be configured and imported back to the cloud before the data can be restored. Avoid this by taking an entire VM system image and using it as a warm VM standby.

4.    Use virtual server backup technology that is designed as agentless from the start.  Instead of requiring that each virtual machine have a backup agent or that an expensive proxy server with snapshot space be provided for backup, some software only requires one agent per physical host that can support all of the virtual servers on the host.

5.    The same rules apply to virtual machines as physical ones. If you back up your physical data to the cloud, why not back up your VMs to take advantage of offsite backups.