As recently as five years ago, business continuity and disaster-recovery topics would make IT executives in small to mid-size enterprises grumble that neither was affordable. Now companies of all sizes are realising they can have both using virtualisation technology.
First, let's clarify the difference between business continuity and disaster recovery: Disaster recovery is about re-establishing IT services in the face of large-scale hardware failure or sabotage, facilities failure and/or regional natural disaster. Disaster-recovery capabilities are measured by the amount of time it takes to re-establish services and the amount of data loss. Business continuity is the ability to continue operations with little or no downtime in some of these scenarios.
Affordable shared storage was the first big step forward for disaster recovery for small to mid-size enterprises (SMEs). When data is stored on disks inside a server, it is only as reliable and expandable as those servers: if a motherboard fails, your data is inaccessible, even though the disks may be fine. Storage-area network (SAN) and network-attached storage (NAS) devices pool the storage out of the server, increasing reliability, flexibility and ease of management.
Snapshots and replication, features of the latest generation of these devices, are at the heart of most effective disaster-recovery strategies. Snapshots capture the state of a storage system at a point in time, and replication is used to copy the state of one storage system to another. In conjunction with server imaging software, these snapshot and replication tools allow you to get back up and running on replacement hardware.
Yet there are still two key challenges: getting server hardware on short notice is not always easy, and even with the most recent imaging tools, you probably need the same hardware, which may no longer be available. Matched hardware sitting at a disaster-recovery site solves both these problems, but in most cases is not an affordable option.
Server virtualisation has changed the rules. For one, it improves on imaging software. Virtual machines are fully encapsulated as files and then a thin layer of software - a hypervisor - allows these machines to be restarted on on any hardware. This removes the hardware dependence and makes it possible to use fewer disaster-recovery physical servers to support production systems - allowing more organisations to afford the appropriate spare physical host(s). In fact, often the spare physical hosts have been freed up at the primary site with the consolidation made possible by server virtualisation.
Virtual machine management and site recovery tools provide the interface to instruct the virtual machines to move to the off-site location, where they are restarted in minutes or hours rather than days or weeks that were necessary previously.
With virtualisation, you create a more resilient production architecture - servers are easily moved among physical hosts, shared storage solutions provide enterprise class features with five-nines availability, snapshots provide multiple point in time recovery points - meaning there will be fewer situations when utilising the disaster-recovery strategy will be necessary. Yet, when it is necessary, SMEs can now avoid the costly downtime that can be so damaging for a business.
While business continuity and disaster recovery build a strong case for virtualisation, IT executives are exuberant about the many other benefits.
Imagine an environment of ageing servers with applications on all different servers in a too crowded server room that is a struggle to cool effectively. This is the scenario that Northeast Delta Dental faced before it implemented an optimal server and storage virtualisation solution.
Not only has Northeast Delta Dental built a resilient production architecture and achieved its business continuity and disaster-recovery goals, it also dramatically reduced power and cooling costs and streamlined the management of the entire IT infrastructure.
Data recovery from snapshots rather than tape, is a speedy, simple process. For Dan Kaplan, Northeast Delta Dental IT Manager, the ability to test disaster recovery with this solution is vital - they can do full non-disruptive failover testing during working hours. Further, Kaplan loves the flexibility to add servers as needed for their in-house development, saving time for both the IT department and the developers.
While virtualisation provides the business continuity and disaster recovery that businesses have dreamed of, it can be a bit daunting. A plethora of vendors and platforms exist and not all solutions are created equal. Choosing the best solution for your environment requires more than looking at a list of features and benefits.
It is worthwhile to work with a solutions provider that will get a thorough understanding of your current environment, needs and goals as well as your business service-level agreements in terms of recovery time objectives and recovery point objectives. The type of data and applications in your environment will dictate whether SAN, NAS or a system that supports both will be most effective. Flexibility for future growth should be considered to avoid forklift upgrades.
Once a trusted adviser understands all of this, it can detail an appropriate architecture and recommend the most appropriate vendor equipment to meet your goals. A successful implementation will also require careful knowledge transfer to the IT staff along the way. After all, you have to run the solution once the implementation is done. Even with all of this, questions may arise so make sure that your provider is easily accessible for ongoing support.
This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.
Broadleaf helps organizations leverage the best possible data storage, virtualization, and WAN optimization technologies to accomplish their goals. Providing hardware, software and on-going services, Broadleaf delivers budget-conscious integrated solutions for data storage, data protection, life-cycle management, server and client virtualization and WAN optimization.