In the aftermath of 9/11 WTC attack, some extremely successful businesses had to shut shop while some businesses got up and running the very next day! The only differentiator between them was a well-executed disaster recovery plan. Businesses and organizations, quite obviously, spend 44% of their storage spending on Data protection.
But why data protection and not just backup and restore?
Disasters perhaps might not be daily eventualities but threat of data loss is a perennial nightmare and can strike at will through many factors as below:
Device failure: Disk and controller failure are one of the most unpredictable yet a frequent phenomenon that occurs in data centers. When Fault tolerance techniques like RAID fail, it’s the perfectly restored backup on a replaced disk that relaxes those palm-sweating moments.
Accidental deletion: Production environments are secured with well-defined user and access control lists but instances where experienced administrators or a well-meaning advanced user have accidentally deleted or moved data are not hard to find. A quick snapshot restore of deleted data can be a life saver.
System outage: System outages include power failure, CPU and memory failure or a complete system crash. Any hardware replacement cannot restore system outage completely unless the latest copy of data is restored on it. A disk based restore in case of system/disk failure or a fast failover to a redundant node minimizes downtime and data loss. A quick replay of snapshot might even be able to restore application services in seconds.
Internal/external sabotage: A disgruntled employee, a mole or an evil remote hacker are one of the biggest threat to integrity and security of data. Recent incidents of a rogue employee leaking sensitive data are causing CIO’s lose sleep. An effective data loss protection tool can go a long way in preventing the repercussions.
Various compliance and regulatory directives to retain data for decades have widened the scope of data protection. Organizations want retention solutions that not only reduce storage costs but also allow easy retrieval and more importantly keep data secure and protected. Archiving solutions are an important component in any winning data protection strategy today.
How does one plan a watertight strategy to protect data?
Any data protection strategy should begin by addressing following considerations:
a) Backup and restore of data with minimal user disruptions.
b) Minimal downtime and data loss in event of failure of outage.
c) A location to easily backup and restore.
d) Costs and budget.
e) Ongoing testing of restoration of data.
These considerations are driven by Service Level Agreements (SLA’s) which define Recovery point objectives (RPO) and Recovery time objectives (RTO). In other words, it’s the recovery goals and not the backup capabilities, which should be kept in mind while planning backup solutions. Further, virtualization, cloud and containers have added new dimensions to backup strategy. Backup of not just physical storage boxes but even the virtual machines, hypervisors, containers and even cloud instances have to be an important determinant of any Data protection plans.
Keeping recovery objectives in mind, different media can accordingly be utilized to accomplish the backup and recovery goals. Traditionally, tapes were the preferred medium and are still in use today but demands on speed have pushed them to tertiary position and are being replaced by disks. The backup and recovery time for these are in the range of hours to minutes to even seconds.
Planning a backup solution is a complex process with a high degree of failure possibilities. There is no one-size-fit-all strategy that fits all the customers and scenarios. For small to mid-sized organizations, spending on backup boxes every year or hiring professionals to manage it adds to the capital expenditure which hits even harder in times of credit crunch. They can save this pain and afford some breathing space in operational expenditure too by considering other viable options of backup. Cloud based backups, BaaS and integrated backup appliances not only minimize the outright expenditure but also reduce the pain of managing backup infrastructure.
For larger enterprises, hybrid solutions might be the way to go. Given that organizations have varied types of data, it’s pertinent to deploy a right mix of tools and technologies that caters to different data requirements and effectively reduces costs and administrative overhead. Leveraging storage tiering and De-duplication of data at various levels of storage can substantially reduce storage costs. And as mentioned above, for user related data, deploying cloud or backup appliances directly reduce capex and minimize opex.
Modern data protection goes beyond just backup and restore. It not just focuses on that but also looks at how one can leverage secondary storage datasets to generate value for businesses instead of just lying around. These datasets can be used for DevOps, Analytics, Bigdata processing, and even for testing.
Conclusively, devising an effective backup solution is not a one-time activity but an ongoing process which needs to be continuously evaluated and improvised as data storage dynamics change. But the fundamental question to continue to answer each time is, Can my backup strategy enable me to protect my data and restore it as and when I want? A resounding YES can be the difference between business interrupted and business as usual.