IT TRENDS
Dealing with data disaster
— photo: ANAND PARTHASARATHY
Play it safe: Singapore’s SoftSource Solutions demonstrates the HP Disk to disk to tape (D2D2T) storage backup system.
One of the penalties we pay for an increasingly connected and digital world, is that the speed of doing business has an obverse side to it: the speed at which the core assets of a company in this Information Age – data – can be irretrievably lost.
Data disaster can strike at any time. In an earlier, unconnected age, Nobel laureate Paul Ehrlich famously said: “To err is human; but to really foul things up, you need a computer.”
The Internet has only made his observation truer: Most data outages are the result of system breakdown or inadvertent human error. It is ill advised today to run a computer-intensive business process without putting in place a disaster recovery plan. Yet alarmingly, the regional results of an Information Technology Disaster Recovery Survey released this week, show that one in three Asian companies are ill-equipped to deal with disaster or attack. Half of them put their DR plans to the test once a year or less frequently.
Disaster happened
Thirty per cent of these tests fail; and in the past one year, one-third of the organizations surveyed had to actually execute their DR plan. In other words, the disaster actually happened: it might have been due to hardware or software failure; power failure; external security issues; data leakage either accidental or through malicious employee action.
The survey was carried out by storage, security and systems management leader Symantec-- and the Asia-Pacific component covered organizations with 500 employees or more, in India, China, Japan, Singapore, Malaysia and Korea.
“Failing to plan, is planning to fail” says Basant Rajan, Symantec’s Chief Technology Officer, India. DR is commonly understood to mean the procedures and policies for restoring business-critical operations, including regaining access to data, communications, workspace, and other business processes after a natural or human-induced disaster.
Common techniques include, regular back up of data to tape or disk and sending them to another site; or direct replication at an offsite; offsite ‘mirroring’ of data — as well as more mundane things like surge protectors, uninterruptible power supplies and back up generators.
But any plan must cover all assets of the company — not just the more obvious ones, like applications, data bases and web servers and email, cautions Mr. Rajan. With business increasingly going mobile, does the plan cover workers’ laptops, home PCs, mobile phones?
Mantra virtualisation
Ironically, there are problems for enterprises living at the cutting edge: Virtualization is the new mantra, for both servers and storage: The physical (as different from the logical) entity might reside far away — may be on another continent or on the premises of another entity.
This leads to tremendous savings in system overheads — especially when combined with tools like data de-duplication (saving just copy of an attachment that was copied to multiple email recipients, for example).
Yet virtualization can pose special challenges to the implementation of a DR plan. ‘Native’ tools for data recovery are said to be somewhat immature in virtual environments and the Symantec survey says over one third of virtualized servers are not covered by the organization’s DR plans — a case of ‘out of sight, out of mind’, perhaps!
Why re-evaluation
That is why over half of the respondents have had to re-evaluate their disaster systems because they had to contend with virtual storage or servers.
The answer might lie in special tools that acknowledge that they have to work in a virtual world — like cluster server or volume replication solutions from a company like Veritas ( now a part of Symantec); like HP’s D2D2T StorageWorks back up systems which help enterprise backup disk to disk to another level of tape storage; like NetApp’s SnapVault operator-less remote back up tool; like VMWare’s DRS or VirtualCenter management tools; like Fujitsu’s Eternus storage systems, with built in data recovery strategies….
The tools and technologies are out there: and enterprises which still hesitate at the water’s edge, while sizing up the cost-benefit ratio of putting DR plans into place do so at their peril, will sooner or later be swept away the water when it turns into a flood.
There is no magic bullet to stave off disaster; but a single, holistic disaster recovery solution that embraces both virtual and real environments; remote offices and portable platforms; and one which minimises human involvement will at least mitigate the danger – and the consequences. To do less is to live dangerously — and drag one’s customers into disaster.
ANAND PARTHASARATHY
Printer friendly
page
Send this article to Friends by
E-Mail
Sci Tech
