A 5-point framework for future-proofing backup and recovery

Experts predict that most enterprises will experience a 50x data growth by 2020. Yet many companies are still taking a traditional and generic approach when it comes to backup, recovering and retaining ever-growing amounts of data. The result is sprawling data sets and backup systems, leading to substantial risk of data loss and inability to recover in a timely manner. So how can organizations protect the business data that resides in their mission critical enterprise databases?

Data is the new engine powering the global digital economy, so much so that the insights it unlocks are considered a key source of competitive advantage and are even seen as a form of data capital. Conversely, any compromise of customer-facing databases, loss of enterprise dataor downtime of business applications can be catastrophic to a company.

The challenge is that unexpected outages and the increasing number of cyber-attacks can cause data loss and corruption. It is essential that a company’s mission critical enterprise databases are protected and that timely and accurate data recovery is available to ensure near-zero business disruption.

Yet numerous enterprises are still stuck with archaic data backup and recovery mechanisms. These cause long backup windows, unnecessary overhead on database servers, virtually no insight into recovery capability,unpredictable restore performance and little knowledge as to whether or not backups are valid. The ever-expanding data universe is only making things worse.

So how should companies protect critical databases?

Modernize your backup and recovery strategy

Database-aware, engineered protection

Organizations are recognizing that there are important data protection differences between data types - for example, flat files consisting of single tables of data, commonly used to import data in data warehousing projects versus relational databases. Consequently, they are moving towards specialized, application-aware solutions for critical data instead of the traditional generic, one-size-fits-all approaches. One such example is Oracle’s Zero Data Loss Recovery Appliance. The resulting code level integration with the transactional databases for data protection delivers the end-to-end visibility and automated backup validation, simplifying management and reducing risk to the business.

The results can be quite impactful. Take, GE Aviation, for example. This US based global aircraft manufacturer wanted to create what it refers to as its resilience platform for its new ERP system. The company wanted to reduce the long recovery windows experienced with its legacy system to bring critical systems back online as quickly as possible. As a result, recovery times were reduced from about 20 hours down to 2 hours – a 90% faster database recovery!

To add to the delight, the enterprise achieved an immediate Return on Investment (ROI) by reducing the risk of data loss with a Recovery Point Objective (RPO) of less than a second versus hours. The company also gets peace of mind with end-to-end backup validation without any overhead on personnel or database servers. Effectively, GE Aviation went from “assuming” backups could be successfully restored to knowing current status for all of the protected databases under management.  

In addition to the better system performance, they have also been able to change and improve their entire operating procedure, and the project lead knows there won’t be any 3 AM calls!

Other companies like AMMROC, a world class military maintenance company, achieved 32x faster backups and 8x faster recovery, enabling the company to provide 24x7 maintenance, repair and overhaul services to its customers without delays.KEB Hana cards, a Korean credit card company, achieved 13x faster backups and a sub second recovery point objective, virtually eliminating data loss exposure and ensuring data security.

So, for any new solution, and in order to ensure maximum ROI on your investment, you should consider this 5-point framework for future-proofing your backup and recovery engine.

1. Eliminate data loss – Look for tools and techniques to protect ongoing database transactions with the goal of sub second recovery.
2. Database level recoverability – Architect your database data protection strategy around recovery instead of just backups. For example, insist on automated monitoring and alerting allowing you to know, not guess,the current recovery window and data loss exposure for all databases, not just backup issues. Ensure that the solution provides end-to-end data validation, so you’ll know that you have a good backup and not be surprised by data corruption when you need to recover.
3. Cloud-scale Architecture – Design for ease of management, reducing error-prone processes, with policy-based management leveraging one backup strategy across all databases. The solution should be highly available with no single point of failure and deliver linear performance with scale-out to meet data growth realities.
4. Compliance requirements – Incorporate database level policies with automatedreporting, monitoring and alerts, protecting backups across their full lifecycle including disk backups, tape backups and remote replication. In addition, the solution should meet regulatory requirements for timely recovery - in some cases this can be as short as two hours -with no data loss exposure.
5. Minimal impact backup – Ensure all backup and recovery related processing, including tape backups (if needed) have minimal impact on productions workloads for backup processes such as backups, deduplication, purging of backups, backup validation, or compression.

Already, many leading global enterprises have started to rethink and redesign their backup and recovery strategies around this five-point framework. Is your organization ready to future-proof its backup and recovery engine?

The author is Director & Head - Systems Sales Consulting, Oracle India

Prime Hype DF 2016 EP