Disaster Recovery (DR) is on a rising trend with the mid-market as companies feel the financial impact of IT downtime and realise that a backup solution is not capable of getting them back to service as quickly as they need. 150 of the 350 companies in the World Trade Centre during the 1993 bombing were out of business a year later, highlighting the importance of a disaster recovery solution to business survival.
However, budgets have not met the rising popularity of the concept. IT managers of companies that experienced the 9/11 attack still had to fight for budget to implement disaster recovery plans and technology. A recent report from Spiceworks indicates that budgets in 2015 will remain stagnant. So how do IT Managers and Directors cope with the need to build more resilience into their IT systems without being allocated adequate budget?
Luckily emerging technologies are enabling better resilience at a more affordable cost. Take the ultimate disaster recovery solution to most; an in-house hot standby solution – offering instant failover of your live systems to duplicate servers which kicks in instantaneously without anyone even noticing any impact to productivity. Many companies still feel nervous of hosting their data in the cloud so this enables data to be kept in-house without security and compliance risks. And data can be synchronously copied so there is no loss of data when failover occurs.
An ideal methodology which every company would like to achieve. However, this is a very expensive solution and still limited to those enterprises that can afford the capital outlay of duplicate systems, the bandwidth required for synchronous replication and the maintenance and testing costs. All of which makes this a prohibitive solution for something which will remain dormant for the majority of time.
However, what happens if you take out the capital outlay and the maintenance costs? Could a company offer a standby solution which means you have a duplicate system in a cloud environment that you can switch on when you need it? Emulating the synchronous hot standby solution, while reducing costs is the holy grail of disaster recovery.
Our first in the list of latest technologies that can bring us closer to this Holy Grail is pre-recovery. It works by copying and recovering your entire IT system every 24 hours to a virtual environment. You have a duplicate system that’s been booted, tested and is ready to go when you need it without any further configuration requirements. No capital outlay, just a monthly subscription depending on the number of servers and data quantities you have.
Pre-recovery gives you 100% certainty that your recovered systems will be available in twice the minutes it takes to boot your IT systems. The benefits are that you will get the fastest recovery times and the best protection for your business if the certainty of recovering quickly from IT downtime is your key priority. The limitations are that your recovery point is to the previous evening so if you have a disaster at 3pm then you could lose 6 hours of business data; although RPO’s can be increased with more regular backups if required.
Second on the list is virtual replication. Continuous replication means that every time a virtual machine writes to its virtual disks, this command is cloned and sent to the recovery site. Recovery point objectives are therefore synchronous so there is no loss of data when you need to recover. Recovery times of minutes can also be achieved. The main difference between replication and pre-recovery is that replication is a product and requires your IT team to manage the technology; pre-recovery is a fully managed service that requires no management on your part.
Full testing regimes tend to be annually with replication products, whereas pre-recovery gives daily testing. Replication failures can go unnoticed if you’re not regularly monitoring them, whereas pre-recovery is certified and monitored daily, so any snapshot failures are picked up that day, making it the most reliable solution. RTOs are favourable with replication, and RTO’s and RPO’s objectives can be met depending on what your budget it – there are many replication options to choose from giving you a good deal of flexibility. Ultimately it’s a weigh up between data loss and certainty of recovery as to where your priorities lie and which of these solutions is best.
Third on the list is the cloud. Whilst both pre-recovery and replication utilise cloud technologies, there are many cloud providers out there who offer disaster recovery solutions which give comfortable recovery rates with very attractive pricing models. If a 4 hour recovery time is suitable for your business, then the cloud enables traditional backup companies to add to their service by offering a “backup and build” model. This means your backups will be stored offsite in the cloud, giving you greater piece of mind that they are secure, with the added benefit of having experienced IT engineers available to rebuild and configure your IT systems before restoring your data.
With more of the manual actions occurring post disaster, recovery is less reliable and takes longer but you will get the resource required at your time of need to work through your problems with you. Cloud is worth exploring further but be sure that you understand what they mean by recovery time – is this the time it takes to boot the servers up to restore minimum productivity or the time it takes to get you back to full productivity? Often these two recovery times can have a difference of days.
The market place for disaster recovery is ever evolving and requires some research to find the ultimate solution for you. As long as you know what your RTO and RPO is, and have a budget which suits the reliability of the service you’re looking for then finding a solution shouldn’t be too hard. And with technologies advancing all the time, it won’t be long before instant failover with zero RPOs will be affordable for all businesses.