When businesses are thriving and all information systems are operating as required, Business Continuity Planning and Disaster Recovery (BCP/DR) are the thoughts furthest from a system administrators mind. However, when critical systems fail or become compromised, affecting information availability, questions then come to the forefront around daily operational procedure. How much has this affected our business? How long can we expect to be offline for? What do we do to streamline the efficiency of the recovery process? Questions such as these must be answered by comprehensive business continuity planning and disaster recovery systems implementation, as well as the regular testing of these procedures and systems.
The risks facing Australian businesses
Disaster recovery systems play an important role in business continuity, and protecting data from uncontrollable circumstances should be a priority. For instance, in Australia’s naturally temperamental climate has already seen extensive flooding in North Queensland this year, and with an El Niño based heat wave looming and the national bushfires that come with it, disaster recovery is vital for any Australian business. The repercussions of scenarios such as these are serious. As Australian organisations are forecast to spend AUD$2.5 billion on datacentre systems in 2015, the implications of losing access to, or the ability to run, vital business applications, systems and infrastructure due to these incidents has a tangible impact on business.
The tangible impact of not having a business continuity strategy will depend on the organisation and can be difficult to measure. However, with more day-to-day operations relying on the availability of digital systems, the cost of downtime is likely to be significant. According to the Business Continuity Institute, most (77%) professionals are concerned about the effects of an unplanned IT or telecommunications outage. During the severe flooding in Brisbane in 2011, a number of data centers needed to be powered down as a precautionary measure, further highlighting the need for disaster recovery even if the systems are located in a fault tolerant facility. That event was estimated to cost more than $2 billion and significantly reduced the state’s GDP. A prudent disaster recovery plan is essential to mitigate the risk of downtime brought about by everything from natural disasters to system failures.
Introducing “Snapshot and Replication”
Historically within enterprise solutions, data has been backed up to tapes, hard disks, or remote servers, during which network administrators often see significant performance slow down as well as long waits required for a backup task to complete, limiting its frequency. With the amount of data requiring backup now increasing rapidly, the subsequent man hours associated with traditional data recovery has also become too great. Many must reconsider the efficacy of their backup and restoration strategies, focussing on minimising performance impact, lowering Recovery Point Objectives (RPO), and instant recovery.In its newest incarnation, Synology’s Snapshot and Replication technology embodies these three pillars, providing an alternative to the typical backup/recovery method.
A ‘snapshot’ is a point-in-time copy used to record the whole data status at the time upon being taken. Snapshots use only a small amount of additional storage space, and exert little impact on system performance. It can then be replicated to one or more recovery sites up to every five minutes. So if disaster strikes, businesses can be rest assured that their data is frequently backed up.
Administrators can readily define a frequent scheduled snapshot calendar, with the ability to store up to 256 snapshots for each shared folder. Multiple tasks are possible, with retention policy of snapshots configured by the administrator within respective time ranges: hourly, daily, weekly, monthly and yearly, depending on the nature of the data and how often they are modified. Maximum numbers of snapshot versions to be retained for each time range can also be defined by the administrative policy, with the earliest snapshot version within each time range being retained as standard. It is also worth noting that snapshots can be locked to prevent deletion by administrative defined retention rules.
Retention rules can be customised depending on server location. Primary sites, which may be equipped with faster Solid State Drives (SSDs) can be set to retain fewer snapshots whereas the recovery site utilising standard Hard Disk Drives (HDDs) retaining on a more consistent basis.
Immediate failover and recovery
Disaster level down-time often occurs when least expected, this is where pre-emptive fail-over testing becomes crucial. With the Snapshot and Replication feature, all necessary service recovery actions can be tested after a copy of data has been sent to the destination recovery site, ensuring domain name, file path and application targets will successfully operate in the event of full scale recovery, therefore saving considerable time and money in recovery objectives. For each replication task, a series of features are displayed for the administrator, including connection status and transfer speed along with other vital statistics to ensure that Recovery Point Objectives are achievable across the setup.
In the event that a primary site does fail, due to natural disaster or other catastrophic failure, users can guarantee that data is being protected on such a frequent basis: up to every 5 minutes from a primary site to multiple recovery sites, ensuring the best Recovery Point Objective.
Self-service recovery has also been implemented by Synology for final end users. Once the end user has been given permission at the time of the snapshot, they are then able to access and restore files that were accidentally deleted or overwritten without the input from of the IT administrators, once again reducing workload of admins and helpdesk staff.
Compared with traditional backup/recovery, this approach helps businesses quickly and effectively return to operational capacity through a series of cleverly designed features intended to drastically reduce Recovery Point Objectives and reduce any performance impacts during the backup or restoration process.
The deployment of replication sites is very flexible, with Synology offering four different approaches to suit specific organisational needs. Active/active distributes replicated data to a readily accessible common database, wherein two active servers function as both direct storage and as a backup destination for one other. There’s also the hub and spoke system, which provides several branches all replicating to a central hub; extended replication is essentially a daisy chain of replications, involving replicating site A to be stored at site B, which then replicates again to be stored at site C. Finally, the One-to-Many approach replicates datasets being transferred to multiple sites.
Snapshot and Replication technology enables IT administrators to efficiently and thoroughly back up and replicate existing data through scheduled snapshots, as well as enabling specific retention policies, ensuring greater availability of recovery points and greater distribution of replications to ensure Recovery Point objectives are met.
For medium to large businesses, Synology’s Snapshot and Replication technology is backed by other existing features to offer a complete data protection solution. Data from a primary site can first be replicated to a recovery site, and then backed up to public cloud services such as Amazon S3 or Microsoft Azure to safeguard a company’s digital assets against any disasters. All data can be encrypted before uploading to public clouds. This total solution ensures businesses receive the best of everything – high performance, instant recovery, reliability, and data privacy.
Summary of Key Points
1. Near-continuous data protection for best Recovery Point Objective (replicating data from a primary site to one or more recovery sites － up to every 5 minutes)
2. Quick failover guarantees the lowest Recovery Time Objective
3. Flexible and tailored deployment options - active/active, extended replication, hub-and-spoke, one-to-many
4. Low resources consumed & straightforward to set up