Let’s define DR. Disaster recovery refers to restoring the information and applications that run the business if the data center, servers or other infrastructure become destroyed or damaged. Your downtime tolerance informs your disaster recovery strategy. The term "recovery" implies how quickly the data and applications can be restored.
What's your downtime tolerance?
If your customer’s company lost access to some or all of its data, it’s likely they’d be unable to continue operations until their data access was restored. They wouldn’t know what to bill to their customers, what customers already owe or what they owe to others. Inventory information, manufacturing processes, contractual obligations and competitive intelligence may be unavailable.
The question is not whether or not to implement a DR plan. Depending on the transaction velocity of their business, they’ll need your help to decide what their most cost-effective DR strategy is. With a lower velocity of transactions — usually a smaller volume of larger invoices — it is viable to sustain longer periods of downtime and catch up when the data is restored and systems become available again. But in high-transaction environments, every moment lost may be difficult to recover. The cost of the disaster is also usually proportionate to its scale.
Preparing for quick recovery
Fundamentally, the faster the customer needs data restoration, the more redundancy you must build into their network. If recovery from tape or other media will be too slow, you’ll need to provide redundant storage drives with the ability to failover automatically when the primary drives fail. In higher-transaction environments, you may need to also provide redundant servers. Ultimately, you'll need to explore modern data storage and recovery solutions.
Choose your preferred level of protection.
Combining the best of on-premise deployment and cloud-delivered solutions, you can help your customers balance what they agree to be acceptable downtime with their available budget and the scale of their IT disaster, without compromising the security and preservation of their highest-value data assets.
Data backup and restoration — The most fundamental protection can be provided by simply installing data backup infrastructure on your customer’s premises. Their challenge is they’ll need to have someone rotate media regularly, and monitor to assure the data is indeed being backed up successfully. This is where most simple backup strategies are defeated, as people forget or neglect to perform these functions.
Remote data backup — Perhaps the earliest managed cloud service was remote data backup, a service that predates the popular use of the term “cloud.” You connect your customer’s infrastructure to the remote data backup provider’s via a secure internet connection, and they automatically, regularly back up the data. Benefits include no media rotation, and assurance of protected data with a Service Level Agreement (SLA).
Managed disaster recovery — It can happen. An IT disaster can be caused by hurricanes, earthquakes, fires and other forces that destroy or incapacitate entire buildings, towns and cities. This is where the concept of redundancy becomes critical. Customers may back up data locally, which will be useful should a server or storage device fail. They simply replace the failing device and restore the local copy of their data. But when the outage is caused by a disaster that befalls their building, their area, perhaps even their entire city, they’ll want to be sure their data is replicated far, far away in one or more remote data centers, and available for restoration as soon as they’ve secured a new physical location from which to operate.
Managed continuity of business — Beyond replicating their valuable business data, if your customer cannot afford to stop doing business, you’ll want to replicate their entire infrastructure. When an outage or disaster occurs, their network “fails over” to the redundant data center, and any of their people who are still working just continue to work as if nothing happened. Other users can connect to the secondary data center easily from wherever they can securely access the Internet.
Make the cloud backup case.
The best solution for organizations that have a low downtime tolerance and cannot afford a moment of downtime or data loss, and require full availability of their systems at all times, is to move everything to a cloud provider and let them provide all the redundancy.
It may seem counterintuitive for your customers to think that moving information infrastructure elsewhere is the best solution. Remind them their primary business is not information technology management. Superior cloud service providers like Microsoft Azure, or Amazon Web Services (AWS) have multiple data centers located all over the world, including massive failover capabilities that go beyond providing redundant servers or redundant storage. They provide whole redundant data centers, and redundancy of those redundancies. Since they deliver services to large numbers of customers, they can afford to spread that expense across their customer base.
Since they’ll be dependent upon the internet, they’ll also need a solid DR strategy that involves multiple carriers with circuits that enter their building from separate places. It’s pointless to spend on multiple internet service providers if one backhoe digging in the street outside their office can potentially sever all of their cables simultaneously. Take full advantage of having multiple connections to the internet by bonding them together for better aggregated bandwidth during regular operations, with instant failover should any of your providers have an outage.
All this redundancy may seem like an effort to increase the size and cost of your engagement, but each redundant service corresponds to a time versus budget consideration that you can discuss with your customer and let them decide. Be sure to carefully document every decision they make, so they cannot come back claiming you undersold them.
By following these steps, you'll be equipped to navigate the downtime tolerance vs. budget challenge to successfully support business operations.