More than just cloning data, today’s backup and disaster recovery (DR) strategies involve a wide spectrum such as capacity planning, data distribution, storage performance, storage administration and storage virtualization.
“For this reason, companies should have a clearly articulated and managed vision to help overcome coordination, technology/vendor bias and cost justification challenges,” says Alex Tan, general manager, Asia-Pacific Sales and Marketing, Quantum Corp.’s DLT Group.
“The one fundamental step that enterprises need to ensure is the executive commitment towards business continuity/DR strategies which has to be regarded as a central business process,” he says. “It is important to note that the critical business processes must have availability requirements and business value well-defined by management.”
However, prior to this, companies must see the need to look inwards and to examine their own ability to survive and recover from the unexpected.
Says Steven Coad, general manager for EMC Asean: “companies must realize that 85 per cent of the time, the information needed is not available. This is not the consequence of a disaster but because of planned downtime, which includes backing up or migrating the data, moving the data centre, upgrading the server, testing applications or whatever operational task requires the system to be offline.”
Adds Tom Zack, vice president of Product Marketing, Asia Pacific International Americas, Hitachi Data Systems: “the ability to have a total recovery option available instantly or almost instantly is key for most large businesses that cannot afford downtime for recovery.”
But, it is not that simple.
Lionel Lim, Sun Microsystems’s vice president and managing director, Asia South, says there are several components to providing business continuance which go beyond data continuance. “These components must address application availability. In order to deliver this, a solution must provide compute fail-over, data path fail-over and storage fail-over. Only when all these components are available, will it be possible to provide availability at the application level which is really the problem that customers are trying to solve.”
While data backup and recovery is the basis of all data availability and data continuance capabilities, it does not meet all of the users’ needs for availability and often requires long delays in recovery. That said, it is the most cost-effective way to guard against disasters and loss of data, says Lim.
Then, there is synchronous replication that provides mirroring capabilities in the same array. Host-based mirroring tools extend this capability to work across heterogeneous arrays. Synchronous solutions that work over 200 km are currently starting to appear on the market. For longer distances, asynchronous replication is applicable and mirrors data to a secondary facility that is located remotely, beyond the range of synchronous solutions, Lim says. By and large, with the technologies today, customers can now have their recovery for key applications and systems online and be ready to pick up the workload in the event of a disaster. But, justifying the investment in a DR strategy can prove to be a challenge.
As Coad from EMC points out, “generally, it is difficult for organizations to justify a solution dedicated to DR. This is because in most cases, businesses tend to dwell on the severity of the disaster, how it happened and why nobody saw it coming. It is crucial to remember that it is the consequences not the causes that matter.”
As Eric Hoh, regional director, Veritas Software Asia South notes, the need to invest in DR strategies presents IT organizations with a predicament.
“At a time when economies are sluggish, IT budgets are shrinking and resources are strained, organizations have to decide between making the investment in updating their DR plans and the underlying technology infrastructure,” he says.
Hence, the primary concern will be how to balance the degree of protection with the level of risk and available budget.
Companies must know what data they are storing. To do this, companies have to prioritize and determine what information must be readily accessible as opposed to what can be stored. The most effective way of doing this is by associating quality of service (QOS) rankings to data and applications. And, after a company’s information has been prioritized, it can now efficiently secure its data.
This will also enable companies to feel confident when making heavier protection investment in certain areas of their infrastructure than others.
Adds Hoh from Veritas: “businesses must realize that disasters of any kind and impact will affect their revenue, customer satisfaction and overall productivity. Therefore, it is essential to look to starting a DR plan within their budget, however rudimentary, that can grow as their company grows.”
Nevertheless, Coad of EMC believes that DR sites do not have to be just costly maintenance expenditures or expensive insurance. He cites the example of billings where the faster the organization gets its bills out to customers, the earlier it can collect the money and put it in their books. And, in new application testing, the more frequently and quickly the organization can test its new internal and external applications, the earlier they can roll it out and gain from it, be it in terms of productivity or direct revenue from customers.
