Surveillance data needs to be reliable, actionable, and completely secure. In this white paper, we explain how you can achieve true redundancy and resiliency.
Regardless of geographic location or sector application, the growing dominance of IP, widespread adoption of megapixel cameras, and the rapidly expanding array of video, alarm, and transactional integrations expected of - and made possible by - surveillance solutions, means that protecting data has never been a greater priority.
Download the white paper to discover practical considerations when specifying redundancy and resilience measures, the options available for robust system-level protection, and how effective unit-level protection can strengthen a surveillance network.
Understanding RAID levels and surveillance system suitability
When discussing surveillance system redundancy it is essential to understand the concept of RAID (redundant array of independent disks) and differentiate the pros and cons of differing RAID levels, i.e. the way data is distributed across storage system disks to prevent data loss and/or to enhance processing and performance speeds.
There are seven main RAID levels but, in practical terms, only three are used in relation to surveillance storage and data redundancy – RAID 1, RAID 5, and, increasingly, RAID 6.
RAID 1: This RAID level involves total data mirroring, i.e. exact duplication, on two or more independent disks. Its simplicity is both its strength and its weakness. Making an exact copy of all data on a separate disk means the array will always be fully functional as long as one member drive is operational.
However, because all data is completely duplicated, it can also be both expensive and slow – having to write large volumes of data takes time and uses up a great deal of storage capacity.
In a surveillance setting, where longer retention of higher-definition footage is the norm, RAID 1’s practical and financial limitations start to prove problematic. For large-scale, complex environments that rely heavily on video data, for example casinos, RAID 1 is not an appropriate solution. However, RAID 1 is highly suited to database/application servers as they require fewer disks.
RAID 5: RAID 5 is probably the most common in terms of surveillance storage. Here, data is ‘striped’ across at least three drives with distributed parity (rather than a dedicated parity disk as is the case with RAID 3 and 4). In the event of a single drive failure, this effectively means that all the data needed to prevent data loss can be calculated quickly from the remaining active disks.
Only in the unlikely event of multiple drive failure would data be at risk. Also, because of the way information is spread across multiple drives, performance (i.e. read/write speed) is significantly faster than RAID 1 solutions.
RAID 6: This level is sometimes referred to as ‘double parity RAID’ – essentially because data is distributed and stored in the same way but with the added benefit of a second parity spread across the drives. It is a feature which safeguards stored data even in the event of two drive failures. The downside is that this does mean RAID 6 requires additional storage to create the same storage space and is therefore slightly less efficient than RAID 5.
Until recently, RAID 6 was considered overkill in terms of precautionary measures. However, mirroring the upturn in the use of HD IP cameras, particularly in enterprise-class surveillance solutions where downtime is unacceptable and rapid data retrieval is paramount, RAID 6 is increasingly becoming the new standard.