Managing Data to Improve Disaster Recovery Preparedness

SOURCE: DateKnowledgeCenter.com

The use of tape as the primary backup medium for disaster recovery purposes long ago gave way to disk-based data protection platforms. This approach enables large organizations with massive volumes of data to minimize storage costs, reduce risk of data loss and downtime, retain data online longer, and accelerate backup/restore times.

Managing Large Volumes of Data

In today’s large enterprises with massive data volumes to protect and multiple data centers and disaster recovery (DR) sites to manage, manual data protection is not cost-efficient and does not provide sufficient risk reduction. Large organizations need to back up and move tens of terabytes (often petabytes) of data over a WAN quickly and efficiently. Also, they manage backup and replication policies for hundreds of backup volumes and data types to ensure data is de-duplicated, replicated, stored, and (eventually) securely erased in accordance with strict regulatory requirements.

As a result, most large enterprises are moving to powerful disk-based appliances that enable them to backup data within their backup windows, to store petabytes of data in a single system, and to automate management of their complex data lifecycle policies.

Automation and Integration

For example, one backup application vendor that has pioneered such automation and integration is Symantec through its OpenStorage Technology (OST) plug in for the popular NetBackup backup application. With OST, NetBackup can be more closely integrated with disk-based data protection platforms, enabling enterprises to take full advantage of the advanced capabilities in both NetBackup itself and the backup target. At the same time, enterprise data protection platform technology has advanced to include such innovations as ContentAware byte-level de-duplication and replication that is capable of moving massive data volumes over a WAN with minimal bandwidth for fast, efficient replication. They also include a high degree of automation, detailed dashboards, and support for OST’s Auto Image Replication (A.I.R.), enabling them to be an integral part of the disaster recovery management of all backup data sets. One such platform is capable of backing up 43 TB per hour and can de-duplicate and replicate these volumes without slowing performance.

Together, A.I.R. and advanced enterprise data protection platforms provide the performance, control, flexibility, and automation that enterprises need to centralize management of data protection —from data backup and replication through the expiration and secure electronic destruction of each copy

As the name implies, A.I.R. enables you to automatically backup and replicate copies of data sets without needing to manage multiple catalogues. With A.I.R. the backup is determined by automated storage lifecycle policies (SLPs) enabling enterprises to consolidate data types with different storage plans onto the same enterprise data protection platform for significantly simpler management. Managers simply use SLP to define all copies at once, specifying the storage device and the specific retention for each copy. They then point all the backup policies that follow the same storage plan to that lifecycle.

Synthetic Backup

Another valuable feature is optimized synthetic backup – a capability that dramatically reduces the volume of data that an enterprise needs to backup and replicate. While SLPs may not be necessary for small and medium businesses where manual backup management may be manageable, and tape may even remain an acceptable medium, but in large enterprises with multiple sites, multiple data centers and massive volumes of data, more seamless integration between a robust backup application and a high-performance, disk-based data protection platform should now be considered a business continuity best practice.

A unified set of SLPs combined with storage pooling and multi-tenancy capabilities in the data protection platform are particularly beneficial to large enterprises with multiple business units and demanding recovery time and recovery point objectives (RTOs and RPOs).

Additional advantages of implementing a centralized, highly automated disaster recovery plan include:

• The ability to leverage de-duplication and compression capabilities built into the data protection platform to minimize the size of both master and replicated backup images
• Content-aware byte differential de-duplication to cut the capacity of data to be backed up and replicated without slowing backup performance
• Bandwidth-optimized replication to deliver fast, cost-effective movement of data to geographically-dispersed locations for disaster recovery protection
• Support for active/active, many-to-one and one-to-many topologies to accommodate different business continuity strategies
• Extending a centralized data protection umbrella to remote office locations more effectively and economically
• More affordable consolidation and centralization of a tape infrastructure used for archiving
• An easier way to set multiple, different retention periods in different locations for lower storage utilization and, therefore, lower costs
• The ability to minimize RTO by automating the importing of catalogs to immediately restore mission-critical production applications and systems

Large enterprises should evaluate emerging solutions that can significantly reduce disaster recovery data protection costs while improving recovery times. The advantages of disk-based data protection are clear, especially when specifically designed for the ingest, de-duplication and replication challenges of massive data volumes.