Blue sky thinking: where next for the cloud?
The cloud has had something of an ‘up and down’ journey across a number of different sectors. Some willingly embrace what it offers while others are quick to shun what it can deliver. Suitability for cloud-based services is often the big differentiator, and in the video surveillance market opinion is diverse and conflicting. Benchmark considers the pros and cons of the cloud in video applications.
The cloud is fairly accurate as a description of the potential on offer no matter which side of the fence you sit on. For those who are pro-cloud, it is a defined but remote location which can be common to many users. It is an off-site space that is undeniably discrete from the everyday working environment, but which can be accessed at will.
For those not so keen on such services, the cloud is also a fitting name. It is fluffy, fragile, difficult to define and invariably out of their control.
An argument could be made for either stance, and as usual there is some truth in both sides of the argument. What makes things even more complex is that not all cloud services are the same, nor do they intend to carry any similarities. This means that users have to carefully investigate and understand the service offering and terms of service for each provider they might deal with.
Cloud services can be diverse and many bear little or no relation to others. All that can be inferred when the ‘cloud’ tag is attached to an offering is that a third party server will be used to deliver the procured services. Everything else is a bit fluffy!
Cloud services can include (but are not limited to) file and document sharing, data archiving, connectivity, data delivery and software provision. Licensing models vary from provider to provider, as do capacity, functionality and terms of service.
In a number of markets, cloud services are a great enabler, a cost saver and often a pivotal part of working practices. However, the cloud is not designed to be all things to all men. Cloud services have not been as successful in security – and specifically in video surveillance – as they have been in other sectors. To understand why it is worth considering the cloud-based successes.
Cloud connectivity is an almost expected part of many services in the consumer sector. From data and photograph sharing through to streaming services, basic cloud-based offerings are often free (albeit with restrictions which are lifted in exchange for paid subscriptions). Cloud-based accounts are something of a no-brainer in this sector due to low (or no) costs, despite many users having experienced delays or loss of connectivity at some time or another.
Another area where cloud services have excelled is with regard to Software-as-a-Service (SaaS). This model is based upon the use of software in the cloud, again based upon a subscription fee. What is interesting with SaaS is that the offerings have generally evolved from being a could-based server hosting the software, meaning users had to log in to use it, to more of a licencing and management role, allowing users to download off-line variants of the software.
Often the issues of downtime due to a lack of connectivity are not due to problems with the cloud service provider. As more workers are mobile, and with a wider variety of infrastructure being used, often issues are created on infrastructure between the user and the cloud provider. These are links that typically neither the cloud service provider or the user have any control over!
Interestingly, with the increased interest in smart homes and IoT, the cloud is increasingly being questioned by users as to the validity of the offerings. Research in the smart sectors has highlighted that users want systems which they can fully own and control; there is also a growing unease with third party access to data.
Reasons cited for wanting full control over software, services and data include ‘lost’ data (usually based on experience of problems) and a reluctance to pay for a service which the user could actually carry out themselves. If there is no additional benefit to paying a ‘middle man’ recurring revenue, why do so?
Many cloud service providers are trying to reinvent what they offer, adding value and embracing technologies such as data push. However, a growing number of end users are recognising that despite the promises and best will of these providers, they cannot control the infrastructure between their cloud servers and the user’s systems. Any failure in a WAN might result in the user losing access to data or services, and where continuity is vital that’s not something they’re willing to pay for.
The video surveillance sector has a number of fundamental requirements which must be met by any service or software, whether it be cloud- or locally-based.
Firstly, continuity is vital. Systems must be fully operable and accessible around the clock. Downtime is simply not an option.
Secondly, the management of data cannot be compromised in any way. Increasingly security companies are being asked to provide evidence that appropriate steps have been taken to meet data security requirements. Benchmark has spoken to integrators who have been told by their corporate customers that they must personally accept responsibility for data protection if a third party service provider is used.
Thirdly, any hardware used cannot be dedicated to the cloud service. As an example, a few years ago Benchmark tested a camera/recorder device which was accessed via a cloud service. We refused to recommend it on the grounds that if the cloud service was terminated, the costly device would be useless. Less than six months after this, the provider pulled out of the UK security market.
When cloud-based video surveillance services first appeared, the main thrust was towards archiving video. This was pretty much flatly rejected by installers and integrators, predominantly because of bandwidth issues. It wasn’t so much that storing video was expensive (especially HD video which many providers claimed to support but could not – and still cannot – manage), but that few sites had sufficient upload speed to make archiving multiple cameras a reality. Also, with memory being at its lowest price level for many years, cloud storage made little sense.
Little has changed since then: upload speeds are still a significant limitation, memory capacities have increased and costs have fallen, and few cloud archiving services can get close to the capacity that even a small surveillance system requires without costs getting out of control.
Some will argue that cloud services should not be judged as an alternative to ‘local’ systems, but the reality is that the end user will be investing in video surveillance to achieve specific goals. If these goals are in any way compromised, then cloud-based services should not be considered. It is therefore very important that installers and integrators ensure that promised performance can be achieved. Supporters of cloud services can reel out figures and stats that underline just how suited the cloud is to any task, but installers and integrators must take a more pragmatic approach.
It is possible to find cloud video services that, on paper, can match the performance of local systems. It might be argued that these can therefore offer a like-for-like alternative. However, for the vast majority using these systems, infrastructure is often the actual limit to what can be achieved.
If infrastructure falls short of the system requirements, it’s not relevant how powerful the cloud-based servers are or how resilient the service is. It might not be the cloud that is at fault, but the result for the end user will be the same.
It is essential for installers and integrators to understand users’ expectations, and if these are not met the customer is unlikely to be satisfied with any excuses about the cloud.
This does not necessarily mean that the cloud has nothing to offer the video surveillance market. Increasingly, video surveillance systems are not only protecting premises, but also are used to collect business intelligence data, such as people counting, heat mapping and footfall analysis.
These systems will generate reports that add value for the end user, but which don’t always have the same degree of urgency as security- and safety-based video surveillance. It might be inconvenient if a marketeer can’t immediately access a footfall report, but it will not compromise the level of protection on offer.
Footage for security purposes remains on the local system and is not only recorded around the clock, but can be accessed on demand, any time it is required.
While some may advocate carrying out the actual analytics for business intelligence in the cloud, it’s not something that Benchmark would advise. In recent tests, cloud-based business intelligence services were found to be largely inaccurate and unreliable. If the connection is lost, so too is all the data for that period.
However, the data collected by the system is processed and delivered in a report format, and the department which needs the information has no need (or desire) to work their way through a video system to find it.
By delivering these reports to a cloud server, staff can view and interrogate the data without having to interact with the security system. This is a benefit for the security manager and means staff in distributed locations can share and access the business intelligence reports.
In such a scenario, if cloud connectivity is lost, it doesn’t impact on the security system nor will the business intelligence analytics fail. Even if the reports are lost they can be regenerated, because the original data is retained in the local system.
Cloud services are well established in some sectors, but even where they have high levels of acceptance changing customer demands mean many providers are reinventing their business offerings.
Currently, cloud-based services for video surveillance are either aimed at niche sectors or the householder market, and research from the worlds of IoT and smart homes shows that attitudes are changing.
Despite the negatives, evolving service offerings are certainly improving, so the cloud remains a technology to watch!