CIOREVIEW >> Virtualization >>

Key Consideration for VM Data Protection

By CIOReview | Wednesday, May 10, 2017

Data protection is one of the key operations that has been most impacted in the virtual environments. Traditionally, on-premise data centers or physical storage infrastructures usually follow the method of restoring the system for data protection. However this approach is not applicable for VM data protection due to its consolidated architecture. Physical systems treat each machine as a primary unit of backup and can be rolled back to a previous point in wake of a security issue. The entire requirement changes in the VM landscape as multiple virtual servers share the same physical server. Therefore it is not advisable to make changes to the server for the need of one virtual server that affects all other virtual servers. VMs now share resources across multiple VMs with each running different applications of their own. There needs to be an effective way to deal with the VM data protection methods. The challenge lies in the building or making changes to one VM leaving the rest of them running without disruption.

The most required capabilities in this regard would be to gain visibility into the files or each virtual server’s logical unit numbers—to surgically separate them and manipulate individually for desired changes. These careful operations require specialized tools capable of offering granularity and unmatched visibility like Veeam—a solution provider that delivers high-speed recovery, data loss avoidance, verified recoverability, leveraged data and complete visibility, to control and restore at precisely individual VM levels. With increased demands for round-the-clock availability of technology coupled with the non-disruptive capabilities in moving virtual servers to a different functioning platform in the event of an outage or failure, the potential for true high-availability failover emerges to be more realistic. In such situations, leveraging on the virtual server’s capabilities to shuffle machines in according with the traffic and performance needs in addition to navigate around failures without user impact will be best practice. This is in contrast in the case of on-prime server infrastructures. Further, backing off from the methodologies practiced to protect physical server environments serves best in the case of virtual server environment. Organizations need to examine the business needs primarily considering down time and data loss tolerance before building a backup infrastructure.     

Agent-free backup is another method to protect VM data. The software mentioned as agent here is installed into the VM to assist backup processes. The software features application-awareness that effectively allows granular backup and database recovery or emails. This approach also accelerates raw backup performance—though only in some cases. Solution providers also offer helper applications that can be coupled with agent-free backups to enable scanning, searching, and extracting granular components from various data types including Microsoft Exchange, SQL Server and Oracle.  

Above all, take in to consideration, the data size and criticality of the VM data with respect to the hardware and software compatibility. Sometimes, an SAN-based VM backup system might be an appropriate solution for particular workloads which may not be fitting properly for a system that needs a VMware Consolidated Backup. Further, by selectively backing up VMs based on relevancy can avoid latency and accurately balancing the tool’s capabilities as per data needs can create improved disaster recovery plans.

Bottom Line

Today, there are multifarious ways to protect VM data with constant upgrades and innovations from VMware. The innovative solutions offered exhibit multiple capabilities that also eliminate the need for separate business continuity and disaster recovery applications. Enterprises while selecting apt solutions for their infrastructure need to carefully choose from a cost and complexity standpoint with the right set of tools and process options. Further, investing in terms of time to clearly understand the benchmark requirements for VM data protection will offer insights into details of previously built systems to make appropriate changes as per requirements.