The impact of virtualization on the IT Infrastructure has shaken IT at its core. Virtualization is changing the current model of assigning one physical server for each application by effectively consolidating multiple servers onto one piece of server hardware and then optimizing its resources. Despite these benefits, virtualization initiatives can prove challenging. For example, deploying certain components of the IT infrastructure with backup and recovery is one area where there can be unanticipated challenges in the support of virtual infrastructures.
According to the article, Virtualization Shakes Up Backup Strategy, traditional backup systems provide “tried and trued support for storage-area networks (SANs), fibre optics, and the latest operating system and server hardware updates.” However, “they are not geared specifically for the complex world of virtualization, which involves multiple guest operating systems on the same box.” In the same article, Dave Russell, vice president of research for servers and storage at Gartner, outlines three of the most popular strategies for virtualization backups:
- Put software agents on each virtual machine (VM) and then use traditional enterprise backup software;
- Create an image of the VM and either use a storage service hosted elsewhere or take daily snapshots of the logical unit number (LUN);
- Use VMware consolidated backup (VCB) that incrementally archives the VM by copying what has changed since the last backup.
These traditional backup solutions may accomplish the task at hand; namely backup and recovery by bringing over multiple backup solutions to a virtual environment (one for databases, one for Exchange, one for SAP, etc.). But they degrade performance on the VMs and create significant overhead on the underlying server hardware. This undermines the very reason that organizations adopted virtualization in the first place, which was to consolidate and simplify management of heterogeneous environments.
The REO BCA takes a solutions-centric approach to address the multiple requirements that organizations have surrounding Business Continuity (BC) and Disaster Recovery (DR). This means organizations can use a single solution to protect servers in their main data center as well as in remote and branch offices (ROBOs). This comprehensive, singular approach to backup and recovery, regardless of where the VM and its application reside, resolves problems on two levels in virtualized environments.
First, the REO BCA provides an environmental profiler to analyze existing applications, data, network, storage resources and utilization. This gives administrators a powerful reporting tool to understand how changes to application data on individual VMs impact the host server and the network. It also serves as an invaluable tool to help optimize resources to better meet BC and DR requirements for applications on each VM. This reporting across the virtual environment is key to building sound BC plans, optimizing traffic over LAN/WAN connections as well as meeting RTOs and RPOs for both local and offsite recoveries.
On a secondary, deeper level, the REO BCA provides lightweight drivers on VMs that may be located either locally or remotely. These are the same agents previously used to analyze the environment but are now repurposed for replication. They streamline the data replication process by continuously monitoring the application on that VM and then replicating every data change at the block-level back to the BCA. By replicating all changes, normal backup windows are eliminated and the risk of data loss that can occur between each backup is minimized or eliminated.
What is most notable about the BCA is that it supports granular recoveries down to the application level. Because it integrates with today’s mission-critical applications (e.g., Microsoft Exchange, SharePoint and Oracle), BCA can understand the state of these applications as replication occurs so application consistent restorations can be performed if needed. Administrators can simply tag an “event” to create a form of bookmark that then can be used later to perform a fast recovery or failover operations. What makes this functionality unique is that it gives organizations a high level of assurance that the application will come up correctly.
The beauty of the BCA is that recoveries are both application and crash-consistent to any past point in time, so organizations can leverage the BCA to first protect their data and then leverage it to achieve near-zero recovery time objectives (RTO) and recovery point objectives (RPO).
It has been reported that backup success rates are improving even as recovery remains a major concern with only 28 percent of respondents feeling comfortable in their backup/recovery strategy. Migrating to a virtual environment will only contribute to continued erosion in these already low comfort levels unless organizations begin to change how they protect and recover their data. Consolidating machines, applications and databases within virtual environments make sense, but bringing existing and antiquated backup strategies along only serves to strain host servers and produce large amounts of uncontrolled data traffic across the corporate LAN/WAN.
The good news is Overland Storage has stepped up and developed a solution that incorporates application awareness, local backup and remote DR into a single solution for virtualized environments. Overland recognizes the pitfalls of traditional backup methods in virtual environments and now offers organizations an important new way to counteract these downfalls. The new BCA is a solution that solves current backup problems in virtualized environments while simultaneously resolving emerging business continuity and disaster recovery challenges as well.