|
The continued exponential growth of data and longer retention requirements have created a new set of challenges for IT managers and business owners. More data created and stored translates into more data to be protected. When storage requirements are combined with the push for greener IT, the problem no longer can be solved by adding capacity - reassessment of the data protection architecture, technologies, and processes must take place to achieve desired results.
According to an IDC study, the amount of data created and stored has increased fifty-fold over a three-year period. At the same time, backup windows either have not changed or have contracted. The explosive growth of digital data is occurring in organizations of every size. In addition, increased regulatory and legal pressures have contributed greatly to the upsurge of data that is maintained for an extended period of time. Enterprises are struggling to answer questions such as, "What happens if an outage causes data loss?" and "How can an organization maintain regulatory compliance and business continuity in spite of data growth?"
The expansion of operations globally, increased online customer communications, and use of offshore resources have put pressure on IT to decrease planned and unplanned downtime and deliver more stringent recovery point objectives (RPOs) and recovery time objectives (RTOs). In the event of a disaster, there is tremendous pressure to recover data, and the processes that depend on that data, as quickly as possible. Organizations are beginning to realize that they will be faced with a data protection crisis unless they take proactive steps in addressing evolving challenges. The deluge of data in need of protection, the shrinking backup windows, and the emerging complexities in storage environments are driving managers to consider new approaches to data protection and management as a way to tame the data beast. Additional demands for granular recovery of critical applications have forced IT managers to consider emerging technologies that deliver near-continuous data protection with smaller intervals between recovery points. Though business owners demand faster recovery and fewer downtimes, they are not willing to spend significantly more and often look to do more with less. Activities such as consolidation, technology refreshes, and implementation of tiered storage are initiatives that are driven by costs. As a result, organizations are looking for solutions that simplify data management and enable businesses to ensure operational continuity, data protection, and compliance and litigation support without putting strain on the bottom line.
Tape Versus Disk Solutions
Unfortunately, the costs associated with replacing existing backup systems or updating infrastructure are forcing enterprises to look for solutions that leverage existing assets and deliver the greatest impact for every dollar spent. Tape-based systems have served as the primary target and repository for data backups, but emerging challenges are driving many managers to consider augmenting tape with disk systems. There are significant inherent issues with physical tape related to performance, recoverability, and reliability. Tape-based environments often experience delays due to robotics, risk failure as a consequence of lack of redundancy, suffer from the effects of "shoeshining," and expose data to security breaches at the time of manually transporting cartridges offsite.
Disk-based backup for digital data frequently has better performance than tape drives, which rely on the physical movement of the recording medium or the reading head. "Shoeshining" is what can happen when a linear tape drive runs out of data kept in its buffer and has to slow down or stop and wait until more data arrives. Today’s tape drives spin very fast; having to slow down or stop takes time, causing the head to jump too far ahead on the media. Before resuming writing of data, the head must rewind the tape and reposition itself. This forward and backward motion is referred to as "shoeshining" and causes degradation in performance and reliability (through wear and tear) of the drive.
Unlike tape drives, disk drives are random access media and don’t succumb to performance penalties as a result of inconsistent data flow. Backup streams can write to disk as fast or slow as possible. Aggregating data through multiplexing can make restoring data via the tape drive even more time consuming; jobs can become fractured and written in segments intertwined with other jobs along the tape, requiring numerous stop, forward, and read motions to recover a single file or data set.
|
