Datacenter downtime means an organization can’t serve its customers and it can’t execute transactions, leading, potentially, to thousands of dollars of lost revenue. Recent data shows that 76 percent of companies experienced an outage in the last year and only 13 percent of those outages were the result of natural disasters. Most “disasters” are related to human error and minor power outages, and the expectation is that data recovery should be fast without flooding, torrential rain, or other factors to address. However, using a backup solution to meet business continuity needs will not work. Data backup is simply not a comprehensive information and application recovery solution.
Common examples of backup methods include off-site tape and cloud storage. Many companies think these methods will protect them if there’s an outage or a disaster. The unfortunate reality is that while backup is generally inexpensive and convenient, it does not ensure quick recovery when a disaster occurs – it only ensures that the data is stored somewhere and can be accessed – eventually. This means that after an outage, a business will have a considerable delay in getting up and running (also known as the recovery time objective or RTO) trying to piece together data and applications so that operations can continue. In addition, backup offerings are often based on older technology, hence their inexpensive price-tags.
To get more specific, let’s review some key attributes.
Recovery service levels. A simple way to define service levels is through RTOs and recovery point objectives, or RPOs. As suggested, recovery time objective is the duration of time within which a process must be restored to avoid a break in business continuity. Recovery point objective is the age of files that must be recovered from storage for normal operations to resume – or the data a business is willing to lose. Backup products do not deliver short RPOs and RTOs because backups usually happen once per day, often at night after normal business operations, meaning a business could be losing a day’s worth of work, in addition to having a lengthy delay for the environment to be restored. This is unacceptable for most businesses.
Application impact. With backup, making a copy of an application and its data drains the CPU on the server. This means even if taking backups more often is an option it still adversely impacts end-user productivity. Additionally, the backup window is a fixed period of time, which is almost always overnight. This strategy does not allow for scalability as applications and the needs of the organization grow. Furthermore, an eight hour window may not be enough time to complete the backup.
Retention. Backups are typically stored for a very long time for compliance and audit purposes, which takes up storage space. Additionally, with backup, the organization will have only one snapshot of the application and its associated data.
Manual recovery. Building the environment from a backup, especially a tape backup, is an extremely cumbersome and time consuming process, stretching RTOs beyond acceptable timetables. Ask just about any IT system administrator about re-building an environment from a tape backup, and you’re likely to get a horror story the likes of which you will not want to endure.
Reverse replication. Using backup, when an application is available at a secondary site (if the backup is being replicated), end-users can use it. However, that doesn’t mean the application continues to be protected. A backup solution will not start taking backups and ship them back to the primary datacenter site.
What these dangers suggest is that the datacenter has changed, and a backup solution alone cannot meet the needs of today’s virtualization infrastructure. The solution to this challenge is enterprise-class disaster recovery (DR) and the best way to implement it is to make it part of the virtualization strategy.
An enterprise-class, virtualized DR strategy offers RPOs of seconds and RTOs of minutes. These are service levels that mission-critical applications can tolerate. With an enterprise-class DR solution, an organization can protect an entire application spanning multiple virtual machines (VMs), servers and storage arrays with a single policy and a single automated failover and failback process. Additionally, DR information is stored for hours or days, allowing for many points in time available for recovery. A virtualized DR solution will also ensure the replicated application is protected after the failover by replicating back to the source site.
Replicating the data is only half the solution. The DR process needs to be executed, in a timely fashion, without error. A comprehensive disaster recovery strategy must include automation to eliminate as many opportunities for error as possible and accelerate recovery, reducing RTOs. Non-disruptive testing further reduces errors. It is critical that testing can be done without impacting the applications and data so that end-user productivity is not affected in any way. With virtualized DR, the environment is protected and testing can be stopped immediately, and a live failover can be executed if an event occurs during a test. Once testing is complete, organizations can see that failover and recovery and failback will perform as the business requires. This test environment can be leveraged to meet other business needs as well, such as test and development.
Ultimately, backup solutions will not get an organization up and running quickly after a disaster – to sleep better at night, an oragnization needs an enterprise-class, virtualized DR solution in place. This kind of comprehensive, virtualized solution – with automated failover, failback and testing – provides peace of mind for organizations that their applications are protected. Most importantly, it means customers will have no interruption in their user experience following a disaster or disruption and corporations are able to protect not only their data, but also their revenue and brand.
The author is solely responsible for the content.
Meet Boston's coolest, smartest and most dynamic founders in our REEL Innovators video series!