System downtime can mean a downturn — in revenue. Remote data mirroring can save you both headaches and money.
In today’s volatile business climate, data is the key to survival. Behind the viability of any successful organization, big or small, is the guarantee of the ongoing flow of information. If and when that flow comes to a halt — even for a short period of time — the effect will be critical.
More than ever before, businesses depend on information availability and access. For most companies, losing access to their data for a prolonged period of time will result in serious consequences and the cost of systems and information unavailability can amount to millions of dollars per hour.
These downtime costs can add up at a truly staggering pace. In a one-month period in early 1999, at the height of the dot-com bubble, online trader E*Trade suffered four system outages. The result was a 22 per cent loss on the company’s stock price. In June 1999, Internet auction firm eBay was hit by 22 hours of operating system failure. The costs? A revenue hit tagged at between $3 million and $5 million, and a 26 per cent decline in the company’s stock price.
Continuous system and data availability is arguably the most critical business and organizational requirement today. Fast recovery from failures and disasters is a logical corollary.
Uninterrupted availability is most commonly achieved by the presence of redundant design, clustering and fail-over with software and firmware support. The bottom line is that IT managers must ensure all critical components are duplicated, so that system and application availability is maintained in the event of any single component failure.
For a speedy end to system downtime, and to facilitate resumption of business in the face of disaster, the most common facility is the use of remote data mirroring. This means that an exact copy of the business’ production data is maintained in a remote site. Should a failure or disaster occur in the original production site, remote mirroring can be used to restart the system and applications at the remote site.
Remote data mirroring can take one of two forms: physical mirroring, based on a replication of the original site’s hardware, and logical mirroring, based on a replication of the original system’s file structure.
Each has its own advantages and benefits, and each will be chosen according to the specific needs of the user.
Physical mirroring will be the appropriate choice when performance, data currency and ease of management are most important factors for the user, the latter being particular important for smaller organizations with a limited IT staff. This is because physical mirroring’s use of a disk-based system does not consume the host systems CPUs. It requires that only a single I/O be issued for the mirroring operation, enabling software mirroring independent of disk technology that may improve read performance with multiple read devices.
Logical mirroring, on the other hand, will be the more appropriate choice when transactional consistency is a more important factor. Within a logical mirroring solution, remote data corruption is less likely to occur, resynchronization will usually require manual intervention, transactions – and not data blocks – will be mirrored, and the result is slightly lower performance than with a physical mirroring solution.
When considering the choice of a remote mirroring solution, a number of questions need to be answered, so that the results obtained will fit the needs of the organization in the most suitable manner.
How important is data currency in the backup and restore functions — the answer will determine whether the mirroring should be synchronous or asynchronous. What is the system’s (and the organization’s) tolerances to some measure of data loss? Should the solution have the ability to maintain changed data information if a second fault event occurs? How can the recoverability of data at the remote location be guaranteed?
If the ongoing availability of data is the key to business survival, a remote mirroring solution — whether it’s based on a physical or logical system — can literally be the guarantee that essential information can be accessed when needed.
Bill Margeson is president of CBL Data Recovery Technologies in Armonk, N.Y.
>www.cbltech.com<. Contact him at firstname.lastname@example.org.