Disaster recovery has always been that insurance policy that gets short shrift. Yet with virtualization coming on strong and injecting new layers of complexity, it's incumbent for IT to revisit disaster recovery strategies to ensure they are up to the task of protecting business-critical services.
A comprehensive disaster recovery plan needs to account for both data and application recovery across both physical and virtual worlds. However, Symantec's confirmed that many virtual systems are still not adequately protected. Almost half of respondents (44 percent) said data on virtual systems is not regularly backed up and only one in five respondents used replication and failover technologies to protect virtual environments. Respondents also indicated that 60% of virtualized servers are not covered in their current disaster recovery (DR) plans.
Part of the problem stems from the lack of visibility across the virtual and physical infrastructure. Many organizations deploy two different backup and restore infrastructures, one for physical machines and the other for virtual machines. Not only does this approach create undue complexity by introducing another product to manage, it also promotes a siloed view of the infrastructure, which can leave mission-critical applications and data unprotected.
IT organizations also appear to be stuck trying to make traditional backup and restore practices work within a virtual world. For example, most organizations have historically put a backup client in their physical server for protection, but that approach isn't effective in a virtualized setup. Because both backup and virtualization are CPU and I/O-intensive processes, the traditional approach can choke the network and degrade application performance, which in turn, impedes IT's ability to meet critical recovery time and recovery point objectives.
In light of the challenges brought on by virtualization, IT organizations should reevaluate disaster recovery strategies and keep the following in mind:
Deploy integrated tool sets. Beyond the acquisition costs associated with buying separate physical and virtual disaster recovery tools, there are added expenses around maintenance and training, not to mention soft expenses related to the complexity of managing separate tools. Employing an integrated tool set that spans both virtual and physical environments provides more standardization and consistency around disaster recovery policies and SLAs. It also eliminates possible confusion and conflicts between traditional data back up and virtualization administration teams, which in many cases, are not the same entity.
Take a unified data protection approach. Companies require a VM image or VMDK for disaster recovery purposes, but Symantec research shows that 80% of restore requests are for individual files and folders. The traditional way has been to sacrifice one backup procedure for the other or go through the pain of conducting both processes using different tool sets and incurring the cost of maintaining twice the amount of disk storage and manpower required to oversee both operations. Moving forward, companies should evaluate their disaster recovery solutions with an eye towards having the visibility and flexibility to restore single files and the full VMDK image for disaster recovery from a single backup pass.
Efficient use of storage. Visibility into the virtual machine and the backup stream can also help optimize the use of deduplication to minimize storage requirements across both the virtual and physical infrastructure. Deduplication ensures that identical, redundant data isn't moved between disk, tape, or cloud storage, and to be most effective, the capability should be part of the disaster recovery solution rather than a software add-on.
Migration support. Having some level of support for migrating physical servers into the virtual infrastructure or between two virtual machines is a big plus for a disaster recovery solution. Not only does it help an organization accelerate virtualization, it provides a cost-effective, automated way to migrate images to lower cost disk storage or to replicate them to a secondary site, which can eliminate the need to transport tapes to costly, dedicated disaster recovery sites.
Given the fast pace of business today and society's near zero tolerance for down-time, what company can afford to put disaster recovery requirements on the backburner? Being able to recover data and business-critical applications effectively and within established recovery time and recovery point objectives can make or break a company, and IT organizations must take the time to evaluate whether their DR solutions can hold up given the added complexities brought on by virtualization.
Symantec's Backup Exec and Net Backup, enabled by V-Ray technology which delivers visibility into virtual machines and applications, can help break down the barriers between virtual and physical silos. Find out more about the V-Ray technology and how Symantec offerings deliver a comprehensive backup and disaster recovery solution.