With the proliferation of video surveillance cameras, the amount of video being created for physical security is growing exponentially. In addition, regulations require organizations to archive all of this video for longer periods of time. These factors make video surveillance archiving a critical issue from both a business and a technology perspective.
Cloud storage has made it easy to store large amounts of data. As a result, many organizations are evaluating a cloud storage strategy for their video surveillance data for its perceived convenience, scalability, and other benefits.
But one factor that must not be overlooked when deciding on a storage strategy for video surveillance data is how the selected strategy will impact video availability. Organizations would be wise to remember that data loss and data corruption problems do occur, and no company is exempt.
Unfortunately, some organizations are tempted to take a simple – and potentially expensive – approach to storing their video surveillance data. They’re so busy running their business operations that it’s hard to find time to research the nuances of different storage. “Just stow it,” they say, seeking convenience over complexity, and believing it’s unlikely they’ll ever have problems accessing their archived video.
The Ugly Reality of Video Loss
While organizations can prepare for many things, video loss is something that can happen suddenly and unexpectedly. While human error – whether by accident or an intentional act – is a common cause of loss, natural disasters or a sudden surge or loss of power can also cause data to be lost in an instant.
If any business needed a reminder of the importance of data protection and archiving, they got it with ransomware. Over the last few years, ransomware has caused lengthy outages and disrupted government operations, hospitals, educational institutions and more. It has even forced some organizations and medical practices to close down for good.
The ransomware trend is not slowing down. According to a New York Times report, ransomware attacks were up 41 percent in 2019 and ransom payments in the last quarter of the year averaged $84,116—increasing to more than $190,000 in the month of December. While these numbers are shocking, ransom payments are only one aspect of this threat.
Ransomware is malware that infiltrates computer systems, often through infected email attachments, and encrypts data files. When the files are accessed, the user receives a ransom message that informs them that their files are not available and that they must pay a fee to receive an encryption key to unlock their files. If they don’t pay the ransom, their files will remain inaccessible, and their computer systems will shut down.
The video surveillance industry has had its own share of high-profile ransomware incidents. One of the most well-known was the infiltration of surveillance cameras in Washington, DC just before the 2017 presidential inauguration. Another high-profile incident occurred with a police department in a suburb of Dallas, TX that lost eight years of video after refusing to pay the ransom demands.
Why is ransomware so effective? Considering the total cost of an outage—which reached an average of $3.92 million globally in 2019, according to the Ponemon Institute—paying a fee to get back in business seems to make sense. The fact is many companies find it less costly overall to pay a ransom demand than to go through the agony of trying to restore their data. But it doesn’t have to be that way.
Know Your Data, Protect Your Data
Knowing your data is an important factor in selecting a storage strategy. As the digital universe grows, so too does the volume of data that must be managed - but not all of that data is critical, and some data actually loses value over time.
One reason many organizations struggle with video availability is that they have never classified their video. Classification entails categorizing video and prioritizing its importance to the organization: simple video classification enables a business to distinguish between critical and non-critical video, as well as prioritize essential video based on its lifecycle (hot, warm, or cold). Instead of treating all video the same, video classification combined with well-defined video protection schemes enables a customized storage strategy to be implemented that optimizes storage cost and performance based on the needs of the business without jeopardizing video availability.
Data Immutability
Another factor to consider when selecting a storage strategy is data immutability. If critical video surveillance data has been damaged or corrupted by ransomware (or some other nefarious act), how does an organization know that any copies of the video or any video archives aren’t also corrupted?
With data immutability, data files cannot be corrupted or modified once they have been stored on media. It can be achieved in several ways, depending on which storage components are deployed within the strategy. But regardless of how it’s done, data immutability is essential to ensure the sanctity of video copies or archives.
Remember, hackers intent on causing harm are creative and persistent. Anything connected to a network is at risk of being breached.
Four Storage Options for Video Surveillance Archiving
When selecting a storage strategy for video surveillance archiving, there are benefits and drawbacks to every option.
Cloud Storage: Archiving video surveillance data to the cloud – public or private – has the potential for many benefits. Cloud storage is almost infinitely scalable, capable of storing growing volumes of data as needed, and organizations need not be concerned with purchasing hardware or building out data center space for storage infrastructure as it’s the responsibility of the cloud provider. The cost to store data in the cloud is relatively inexpensive. For example, pricing for Amazon Web Services (AWS) is $0.023 per gigabyte for standard S3 and $0.004 per gigabyte for S3 Glacier.
But cloud storage has several drawbacks. When data is stored in the cloud, organizations relinquish some control over where the data is located geographically, what devices the files are stored on, and who has management access to it. For some types of data, that isn’t a problem. But for sensitive data or data files protected under stringent industry requirements, storage in the cloud is not an option.
Depending on what cloud storage service is being used, latency – the time it takes to recover data – could be a problem. Available bandwidth will depend on what is purchased from your telecom provider. In addition, if a long-term archive solution is purchased from a cloud provider, such as AWS Deep Glacier, it could take days from the time you request a restore to when your data is re-hydrated and available.
A third and often overlooked challenge of cloud storage is egress cost – the fee to retrieve the data stored in the cloud. These costs can be substantial, as NASA recently learned in an audit report stating that improper modeling of egress charges may have resulted in a significant underestimation of the cost to operate their Earthdata Cloud.
HDD/SSD: Using hard disk drives (HDD) or solid-state drives (SSD) for video surveillance archiving offers high capacity and fast performance. If a redundant site exists, data can be spread across geographies, protecting against a physical disaster scenario. And because of the speed of disk, recovery time is substantially reduced.
But disk also has its disadvantages. Disk is always connected to a network, which makes it susceptible as any device attached to a network is a potential target for hackers. If the archived data is stored on disk, and that disk is corrupted by ransomware, then the integrity of the archive is compromised and cannot be trusted.
There are also overhead expenses associated with maintaining disk. Costs associated with space for the data center, the management and servicing of equipment, and power and cooling can all add up.
Then there’s the cost of disk itself. Although prices for hard disk drives and solid-state drives continue to decrease, disk is still more expensive – particularly high-performance disk arrays.
On-premises object storage: One of the most flexible storage options for organizations looking to safely store video surveillance data is on-premises object storage. This technology saves information as objects and is the best form of storage for large unstructured data sets like video surveillance data. Scalability is nearly limitless, and cost is minimized when data is spread across different nodes.
Some object storage technologies also provide the ability to geographically disperse the video across multiple locations without having to keep full copies at each location. This option protects against losing a full location due to a catastrophe such as a natural disaster in a cost-effective manner unavailable with other storage options.
Magnetic tape: Although it has been around for years, magnetic tape offers many advantages as a medium for video surveillance archiving. Tape provides data immutability so data stored on tapes in a secure facility is not at risk of being corrupted by ransomware or other malware. Tape is also physically attached to a network, creating a barrier that makes that data unreachable to hackers.
Tape also provides high-capacity storage. The current generation of linear tape-open (LTO) tapes, LTO8, can store 12TB raw data (30TB compressed) with data transfer speed of 360MB per second (760MB per second compressed). New generations of tape are in development, and the LTO Consortium—which includes representatives from IBM, Quantum, and Hewlett-Packard Enterprise—has published a technology roadmap that extends out four more generations to LTO12 with capacities up to 360TB of compressed data in a single cartridge.
Magnetic tape is the lowest cost video surveillance storage option on the market today, but it does have some disadvantages if used as a sole archival copy. Tape cartridges, even if not stored within a library, require human handling. To protect against a site disaster, tapes must be packaged and shipped to a secure, offsite facility.
With tape, data retrieval time is slow as the tapes are typically offsite – it takes time to physically retrieve the cartridges and load them. For organizations requiring near-zero availability time, tape alone will not satisfy that requirement – which is why tape is recommended as an archive tier only.
When it comes to selecting a video surveillance data storage strategy to support an organization’s archiving needs, there is no single answer. Factors such as retention and recovery times need to be considered and understanding the relative value of video can be a critical step toward tiering it and storing it using the appropriate storage technology. But with proper planning and understanding of the nuances of storage options, the decision becomes much easier.
About the author: Curt Wittich, VP/Strategic Markets for Quantum.
