DRJ Fall 2019

Conference & Exhibit

Attend The #1 BC/DR Event!

Fall Journal

Volume 32, Issue 3

Full Contents Now Available!

DRJ Blogs

DRJ | The premiere resource for business continuity and disaster recovery

Double or Nothing: Replication Versus Backups

Double or Nothing: Replication Versus Backups

Originally posted by Navisite

The backup strategy used by most businesses has grown organically. In the early days, the systems administrator would manually copy key files to removable media. Initially a stack of floppy disks would have been adequate. Over time, corporate storage capacity grew and the removable media followed – floppies were replaced by high capacity tapes and hard drives.

Back When Point-in-Time Backups Were Adequate

Because the volumes of data in question were relatively small, this approach was perfectly adequate. A simple point-in-time copy of files taken overnight when data operations were on hold was considered sufficient.

Falling costs and the expansion of technological capabilities within storage, coupled with a corresponding increase in the importance of data to business operations created a new problem however. The amount of data being held could not be written to external media quickly enough for the backup job to complete overnight.

The other major problem with traditional backup methods is the time required to recover data. On average it takes three hours to recover from a disaster. Even huge backups can be spanned across multiple media without a problem, but a full system recovery would easily take days. The longer recovery takes, the longer your business will have to operate at reduced capacity.

Secondary Data Centre Replication

For enterprise-class businesses, the answer was to introduce secondary data centres. Using near real-time replication, data could be synchronised to the remote facility, providing a much-needed failover in the event of a disaster.

Initially the technology was incredibly expensive – in effect, you had to buy a complete second infrastructure for key systems – and thus available only to larger companies. And like traditional backups, replication was often done in batches, providing only a point-in-time copy at the remote site. In the event of failure, there was always a possibility that changes between snapshots may be lost.

Despite the shortcomings, early replication was a significant step-up to nightly backups. Mission-critical systems can be brought back online within mere minutes, dramatically reducing the potential impact of a local outage.

The Future is Now – Cloud-Based Replication

Advances in technology have finally opened replication to businesses of all sizes. The combination of low-cost, high-speed broadband and the immense scalability of Cloud services has lowered barriers to entry – not least because organisations don’t need to build a redundant data centre or expend critical budget on a colocation site to host the replicated data anymore.

Modern data replication technologies provide a huge range of options for businesses of all sizes. You can choose to backup physical servers in their entirety, or just the contents of virtual machines for instance. You also have greater control of the replication destination with support for traditional onsite data centres, Cloud services, or any hybrid combination according to your needs.

Cloud-based replication also offers the potential for fine control of data recovery operations. The right provider will allow you to define DR parameters like RPOs, RTOs, as well as resource specifications like processor and memory allocation according to business unit requirements (and budgets). And they will provide your organization defined SLA’s for you to assess their reliability.

Vast amounts of data can be synchronised offsite in real-time – and without significant capital spend.

A Final Warning

Although data replication is available to everyone, it is just one part of a true disaster recovery solution. You must check that information is replicated accurately, and that systems failover correctly when your live site goes offline. Ensure that your data recovery plan includes provisions for testing as you deploy replication systems and be sure to perform the tests regularly – if you don’t, you may be out of luck when a disaster occurs and you need your data back to ensure optimal business continuity.

To learn more about the benefits of data replication, and how your business can take advantage, please get in touch.

Disaster Recovery: what happens when it all goes d...
Boeing 737MAX Issues Illuminate the Costs of Ill-C...