Big Data Storage and Security

In the world of big data, it’s worth remembering that not all data is created equal. There was a time when IT professionals would say “data is data”, but as we’ve become increasingly data-driven, this is no longer the case. Data comes in varying shapes and sizes, with varying degrees of sensitivity and criticality. So, if data is so variable, why would you adopt a one-size-fits-all approach to data management?

Traditional data storage, back-up and recovery systems were more about data integrity than they were speed of access. Some of you will remember the time and resource that went into weekly systems back-up. Yes, weekly. Then there was the monthly off-site backups that involved backing up to tape and having the IT Manager take a copy home for safe keeping.

Today’s IT Managers might laugh at the concept, but 20 years ago, this was common practice. And don’t be too quick to mock, because many of today’s organisations are guilty of the modern equivalent. Whilst the frequency of back-up has increased significantly, some of the fundamentals remain intact; a primary storage array may be replicated to a remote data centre for instant access or backed-up to tape for disaster recovery purposes.

This model may have been suitable in its time, but as data gets bigger and IT estates modernise, it is showing its age. Tape has long been cumbersome as a medium. The need to physically transport tape copies from one location to another exposes your data to a myriad of risks; including loss, theft or damage. At the same time, replication can create complexity and increase management overhead as large volumes of data are copied between data centres.

Data as a strategic asset

As IT’s role has increasingly become one of ubiquitous service delivery, availability is the measure by which everyone is judged. We are so dependent on data for every aspect of day-to-day business, that downtime is operationally and financially significant.

Data has become a valuable commodity. Consequently, it’s also become a more attractive target for cyber-criminals. Since it started its breach level index in 2013, Gemalto estimates that over 7 billion data records have been lost or stolen – that’s an average of over 3,000 per minute.

As we become faster, more agile and more data driven than ever, the demands on modern storage infrastructure are far greater than before. The sheer volume of information generated by today’s big data applications and the Internet-of-Things puts pressure on networks and servers alike.

IT Managers are increasingly looking to hybrid strategies to address the competing challenges of security, accessibility and cost efficiency when it comes to data management. The emergence of flash storage and Cloud computing has had an influence on data management strategies, but technology alone will not provide the answer; there also needs to be a rethink when it comes to process.

A new model for data

Several new models for data protection have emerged over the past few years. Whilst they have their points of difference, they also share a lot of common ground. One thing they all have in common is putting data at the heart of the strategy.

So, when you’re planning your new data management strategy, look at your data from a holistic point of view. Understand what data is “mission critical” and what isn’t. What needs a recovery time objective measured in seconds and what doesn’t.

Match the features of your data storage and recovery solutions to the profile of your data. Where necessary, utilise on-premises storage for high availability and speed of access; or when data must be kept on site for compliance or sovereignty reasons. Although more costly, synchronous replication will minimise downtime and ensure business continuity with access to critical data.
Where applicable, leverage private or public Cloud infrastructure to provide cost-effective scalability, secure data archiving and asynchronous replication as a part of your disaster recovery strategy.

If your data strategy is no longer fit for purpose, you could do worse than start with a business impact analysis. Define your data profiles and plan a strategy that balances price and availability. It might be time to retire the tapes, skip the spinning media and go straight to solid state.

If it’s time to rethink your data management strategy, why not explore the alternatives? Discover more about hybrid storage and Cloud solutions from ONI: