We have developed an off-site backup and disaster recovery process specifically for businesses with multiple servers. This is a comprehensive backup service that stores, monitors and tests the systems, so that organisations can be confident they have a reliable off-site working copy of their systems and information. This windows server backup is ready for disaster recovery. Servers can also be provided ready to recreate working systems in the event of a real disaster, depending on individual client requirements.
The principle operation is to take daily snapshots of the multiple server drives and send these to the off-site datacentre, normally overnight. These snapshots are called images because they contain all the disk data to recreate a computer. The images are automatically put in place and tested daily to ensure their integrity and that they are ready for use.
In addition to automated testing we undertake two forms of manual test, the first is equivalent to the automated test mount of the drives and the second is to start the servers as virtual machines. Starting the servers demonstrates 2 key functions; the first to ensure the server runs and the second to ensure the boot process is practised. This means that in the event of a real disaster we have practised for the event and know the timescale and any additional issues that may be encountered.
The manual mount of the drives is normally undertaken twice a month and the virtual boot of each server every 3 months.
Every organisation will have its own disaster recovery requirements and priorities. We find that most businesses want a DR process in two parts. The first is to have a defined emergency set-up of critical systems running inside a day; the secondary systems can be brought on line over a few days. The requirements are defined by individual clients, in line with their approach to risks, typically flood, fire and theft.
Every server drive will have a daily image; this is part of a chain of images used to recreate the server from the base image. To both simplify, and ensure a faster response, daily images are combined into weekly and monthly images over time. This process is carried out for the complete life of the server, resulting in a monthly history of the server.
For example, if the backup process for the server started 2 and a half years ago (30 months) the data centre will have a base image with 29 individual month images, this will take the server to the beginning of the current month. The daily images from the start of the month will complete the chain of images so that the server can be recreated from the last backup. The daily images are normally deleted once the month image has proven to be correct.
Some organisations like to have simple data files copied and stored instead of (or as well as) the disk images. These files can also be copied off site to the data centre as part of the backup image process on a secure network (VPN).
The secure backup imaging process is based on industry leading software from StorageCraft called ShadowProtect. Some businesses prefer to have their own dedicated equipment in the data centre while others just require a storage platform on shared servers. Regardless of the hardware set up the backup process and service to monitor and test is the same. Some customers prefer to have more frequent manual monitoring and testing, this can be arranged as required.
Where an organisation has multiple sites a backup and DR system can be set up and kept on company premises.