Disruptions are inevitable in business and preparation is key to maintaining continuity. Implementing a Business Continuity and Disaster Recovery (BCDR) plan is critical to ensure your business functions as usual when disasters do strike. To build a comprehensive BCDR strategy, determining your Recovery Time Objective (RTO) and Recovery Point Objective (RPO) is vital. RTO and RPO are two of the most important metrics in BCDR that help quantify the potential losses that ensue if critical systems fail. They are a critical part of a BCDR plan that help in mitigating the impact of downtime and data loss after a disruptive incident.
Debunking RTO and RPO
RTO and RPO are two time-based metrics that are commonly used alongside one another. However, they are not interchangeable.
RTO is a measure of time within which applications and systems must be restored after an outage. It provides a target measured in time to recover IT and business activities. RTO may be determined by how quickly your organization needs to recover as well as what available resources and budget are available to ensure recovery in a timely fashion.
RPO is a measure of the acceptable amount of data loss, prior to a disaster, that a business can endure and still continue to function in the event of a disaster. It represents your organization’s tolerance of loss with regards to data. It is determined by the amount of data in between backups that may be potentially lost and serves as an indicator of how frequently backups are run.
Importance of RTO and RPO
In today’s always-on business landscape, companies cannot afford downtime. But the unfortunate truth is disasters are inevitable and can strike when you least expect them. Any number of incidents, such as cyberattacks, natural disasters, technical failures, power outages or human error, can force you to halt business operations. Therefore, it is important to understand and establish well-defined RTOs and RPOs in order to have a sound BCDR strategy in place.
RTO and RPO help you take the guesswork out of BCDR planning and enable you to prepare better for unforeseen disruptive events. You can align procedures and policies to ensure any data loss and downtime that occurs as result of a disruption is tolerable by the business. This also ensures critical business applications remain unhindered during a crisis or an outage.
Calculating RTO and RPO
Organizations across different industries and sectors have unique RPOs and RTOs, but calculating them usually starts with understanding the cost of downtime. RPO and RTO metrics will differ depending on the criticality of your data and IT systems. For instance, applications that can be down for several hours without impacting your business can have longer RPO and RTO values while client-facing services and applications, which can significantly impact your business, must have more aggressive RTOs and RPOs.
Build an inventory of systems and applications in use by your organization and categorize them into tiers based on their criticality levels. Depending on your business’ SLAs and criticality of each application, the RPO and RTO metrics may be the same or different.
Tips to Improve Your RTO and RPO
Increase Backup Frequency: One way to quickly improve RTOs and RPOs is to increase the frequency of backups. While this may not be feasible for every protected asset, consider more aggressive backup schedules for mission-critical data. More frequent backups will reduce your RPO immediately. Depending on your solution, you may see improvements to your RTO with this method. If full backups are taken more frequently, the chain of dependent incremental or differential backups for a given asset will be fewer between each full, which results in speedier recovery.
Utilize Changed Block Tracking: Accessible via vSphere APIs, VMware’s Changed Block Tracking (CBT) is a feature that aids in performing incremental backups of virtual machines. Backup solutions, including Unitrends, may utilize CBT for any VMs running on ESX/ESXi hosts in order to track disk sectors that have been altered between two changeset IDs. Tracking of block changes occurs outside VMs in the virtualization layer so that when the backup software performs a backup, it can request transmission of only the blocks that have changed since the last backup, or the blocks in use. This ensures short backup windows to meet more aggressive objectives.
If your solution is monitoring for changes to data blocks, by the time the backup runs, pre-processing has already been done, thereby reducing your backup windows as well. In the event of a disaster, changed blocks can be automatically reassembled to a synthetic full backup image (solutions like Unitrends create your synthetic full(s) as a passive background operation keeping recovery chains short to meet shorter RTOs).
Embrace Cloud Technology: You should be replicating a copy of your backups to an alternate target. Traditionally, this may have been physical media such as disk or tape. While legacy media still has its place, your RTO will suffer with this strategy since the media must be retrieved, and the data imported and then recovered. When building your strategy, consider replication to a secondary appliance running in the cloud. This target can be run in either public or private cloud. In either case, your RTO will be based on the time it takes to failover from one host to another. Data replicated to a secondary appliance is kept in a hot state, which is immediately available for recovery.
Implement Synchronous Mirroring: Similar in concept to replication, synchronous mirroring helps provide a zero or near-zero RTO by concurrently writing input/output (I/O) from a primary storage device to a mirrored system. Each time data is written to a local disk, it is also written to the mirrored site. In a mirrored system configuration, the write is not considered complete until confirmation of the write is sent from the mirrored site back to the primary site. This ensures both sites are identical at all times. The secondary copy is stored in a hot state on the target, enabling immediate recovery. This strategy provides high availability across a dual-node cluster.
Improve Your RTO and RPO With Unitrends
Unitrends provides pre-integrated appliances, automated recovery testing and a massive library of supported operating systems, applications, virtual machines and devices that converge your entire backup and recovery needs into a simple, all-in-one solution.
Moreover, our SLA policy automation provides recoverability AND compliance with your RPOs and RTOs. This empowers you to define, analyze and report with testing for DR readiness, depending on your business’ requirements — daily, weekly or monthly.
Listed below are some ways Unitrends can help you improve your RTO and RPO.
Incremental Forever: A backup mode available for File, Image and Host-Level (VM) backups, the Unitrends Incremental Forever strategy runs a full backup one time, followed by incrementals thereafter at the frequency that best suits your environment and recovery needs. The system will synthesize fulls and differentials locally from the incrementals to ensure quick restores. Synthetic Backups are created system-side only and do not impact protected clients or networks. Incremental Forever’s efficiency enables a range of RPOs, regardless of whether your tolerance for data loss is measured in minutes, hours or days.
Image Agent: For use with Windows clients, Unitrends Image Agent (also referred to as block agent) performs backups at the disk volume level, independent of individual file read/writes. Efficient and simple to deploy, the agent utilizes VSS Aware to quiesce applications, such as Exchange and SQL, for application consistency; VSS Full truncates the logs as part of each successful backup. Application Awareness also reduces appliance load for more efficiency since each backup is its own thread. Traditional application protection often protects assets in such a way that each individual database requires its own backup thread. You can recover individual files from image agent backups and you have a number of options for recovery with near-zero RTO, including Instant Recovery (Physical-to-Virtual).
Disaster Recovery as a Service (DRaaS): Enabled by Unitrends Forever Cloud, DRaaS provides rapid spin-up of selected critical systems or your entire data center with just a phone call. Used in conjunction with Unitrends Recovery Series or Unitrends Backup appliances, Unitrends DRaaS provides hands-free business continuity and disaster recovery at a cost significantly lower than building and managing your own off-site DR. Contractually guaranteed recovery SLAs provide RTO options of one hour (maximum time to recovery), 24 hours or Best-Effort. You can mix and match SLAs to build the right response for your organization. Recovery Assurance testing provides regular reports of DR readiness, including RTO/RPO compliance reports and the results of application-level tests to certify recoverability and performance.
Want to learn more about how you can improve your RTOs and RPOs with Unitrends? Get in touch with us today!