Cloud, that vapourous construct of marketers and human imagination, holds many installers, integrators and end users in thrall but they’d feel more relaxed if they simply defined cloud as ‘someone else’s computer’.
PART of the challenge of the transition to a networking landscape that includes cloud is market acceptance. For many people, the word cloud and all those misty images of condensation meant to represent it, are opaque and incomprehensible. When people explain cloud, they seldom mention data centres. Instead they wave their hands mysteriously as though invoking Be'al, the life of everything.
The reality, however, is far more pedestrian. When George Favaloro and Sean O’Sullivan came up with the idea of cloud computing in 1996 they not only began Compaq’s push to supply servers to cloud computing providers but heralded a new frontier of networking. And far from being ambiguous, their ideas revolved around outsourcing local computer services to centralised data centres. Then, as now, the key issues of successful cloud are cost, bandwidth, latency, security and a business model capable of generating revenue.
Over at Bosch, James Layton says cloud has a couple of definitions that need to be considered.
“In the simplest form, it encompasses off-site storage of data (being the storage of data outside of the physical location where the data was created) and off-site processing (being data management or data processing carried out on hardware away from the physical location where the data is generated),” he explains.
“On a more practical note, cloud has also become synonymous with cost aggregation of data storage and processing. Say, I have 1000 users, who all require an equal level of data processing. Under normal circumstances, each user is up for the same hardware costs. But now imagine that at any point in time, only 10 per cent of those users needs to be actively processing data – by using a cloud facility with processing power for only 100 users, but with that processing being shared depending on needs, each user is only up for one tenth the cost.
“Of course in real applications, cloud servers are built to handle higher than the expected load, but there is only a cost saving if the system is built to handle less than the maximum possible load.”
Layton believes cloud will be a significant part of all network topologies in the future.
“Definitely it will be,” he says. “Off-site backup of data has been around in the IT industry for decades, it was just usually handled through the physical movement of storage units such as magnetic tapes. With almost every network now connected to the Internet, the ability to shift data to a secure secondary location has become much easier.
“No many how many backups you do of your system at home, none of that is any good to you if the house itself burns down – off-site storage allows a much greater degree of protection. When it comes to data processing, many users or businesses can’t afford to build large scale server farms required for processing massive amounts of collected data. Using cloud technology, these people are able to share the processing load to offsite facilities and make use of otherwise unused CPU cycles.”
According to Layton, cloud is already a large part of electronic security, and is going to become an even larger part going forward.
“When you look at conventional security hardware, its primary weakness is the physical hardware itself,” Layton explains. “It’s fine to have a DVR in your business to record an incident, but if the perpetrators are able to locate and disable or destroy the device, the evidence is lost. For this reason, you can certainly see the appeal of having the data remotely stored offsite as it is collected.
“Additionally, as analytics become more and more important to electronic security (including features such as facial recognition, left object detection, etc), the choice will either be to place the analytical engine in the field (for instance, at the edge), increasing the cost of the installed hardware, or transferring the analytical intelligence into the cloud where larger and more robust hardware can process the data.”
What are the most important elements of any cloud-based electronic security solution – for instance, adequate layers of infrastructure, the user interface, lack of latency, encryption?
“Cloud services for electronic security need to be a little different than they may be for other IT applications,” Layton says. “Take the example listed above of reducing the cost by only using enough hardware to cover the expected portion of the system required at one time – that’s all well and good until something unforeseen happens and the available load is exceeded.
“In the past, major incidents like earthquakes or mass power outages have caused a large number of alarm panels to report in at once – as a user of security services, you wouldn’t want to think that your data may get lost because the cloud server was too busy processing other requests.
“So, the scale of the infrastructure is a key element – the system needs to be able to maintain a full load. Additionally, like any system designed to provide redundancy, it needs to be robust. More and more data centres are going to pop up in Australia as cloud services become more relevant and choosing the most reliable one is going to become a difficult decision.”
Layton believes that encryption is going to become one of the hardest things to manage with cloud.
“It’s not an issue if a product uses its own proprietary cloud server, but if a manufacturer plans to use one of the larger and more robust providers like Amazon Web Services for their infrastructure, the level of sharing required regarding the encryption of signals is going to be a major consideration,” he says. “Encryption is about the biggest corporate secret a security company can have – the more entities you have to share it with, the greater the chance it is broken, and is suddenly very unsecure.
“One last considering is going to be data ownership. Once you put your data in to the Cloud, you have given someone else a degree of control over it. When you put household items in to a storage facility and then don’t pay your bill – the facility reserves the right to sell the items you stored to recover their costs… could such a thing also be possible with your data?”
What is the greatest challenge for the future of cloud security applications in Australia – is it infrastructure? Is it bandwidth? Is it data security? Is it the security of mobile devices? Is it establishing a viable business model? Is it cost/economies of scale for the end user? Is it misapprehension in the market? Is it something else?
“The biggest problem facing cloud services in Australia currently is the availability of bandwidth,” Layton argues. “Last year in North America the FCC made a ruling that an ISP had to be able to provide 25Mbps bandwidth to be able to use the term broadband to describe their service. Currently, this is the maximum speed we are expecting to see with the NBN for the foreseeable future.
“Without considering expensive fibre links, it is simply not practical to expect to be able to store live high definition footage from your IP camera system in to the cloud. Video for the moment is limited to snapshots, lower resolution, or delayed clippings of video of relevant events, making it difficult to call the Cloud service a truly contemporaneous storage option.
“This isn’t the case for simpler communications such as alarm communications or even home automation control, but at least at the moment, the industry is still finding its feet when it comes to the costs, security and reliability of these services. Looking abroad, in North America basically anything security on the cloud is going to cost you, whereas in most European countries, a lot of security manufacturers are offering cloud services for little or no charge.”
What do you see as being the greatest strengths of cloud? For instance, storage redundancy, redundancy of power, removal of the need to maintain network hardware and a secure network space on site, reduction in costs to power and cool network hardware, delivery of powerful remote capabilities to simple local security systems, the ability to allow remote management of systems, etc?
“There is actually a very immediate benefit of cloud services today, and that is the fact that it is helping the security industry adapt to IP-based infrastructure,” Layton explains. “Technology has changed over time, and one of the biggest struggles a lot of veteran system integrators are dealing with is learning the basic of layer 3 networking – the need to forward ports for a device to communicate outside its own network. Most cloud based products now provide port tunnelling services that remove this need and are helping less IP savvy installers keep up with the technology.
“As cloud becomes more commonplace, the initial consumer draw is likely to come from the remote access and control functionalities. These days everyone owns a smartphone, and having the ability to quickly check the cameras at home to make sure the kids have gotten home from school is going to be the sort of feature that appeals to the end user, and gets them across the hurdle of the additional cost of cloud. As global manufacturers push ahead with The Internet of Things, connectivity, or the perception thereof, is going to be a key driver in purchasing decisions.
“Getting deeper in to the security side, the true benefit is going to come from the inherent redundancy offered by cloud, along with the enhanced capabilities we should start to see in electronic security systems as manufacturers come to terms with the possibilities of remote data processing.”
Seadan is the largest distributor of Risco alarm panels and detectors in Australia and over the last 18 months the company has embraced cloud as a back end to provide remote access and management of these systems.
According to Risco’s technical manager, Peter O’Callaghan, the concept of cloud combines infrastructure, hardware and user interface.
“In terms of Risco’s cloud, it’s a global network of servers that are interconnected to provide a stable secure platform to allow customers to access their Risco alarm panels from anywhere using their smartphone app, Web UI, or configuration software,” explains O’Callaghan. “The cloud allows the easiest way to install and configure network-based systems, and allow reduction of cost and increase in reliability.”
O'Callaghan believes cloud will be a significant part of all network topologies in the future.
“The trend is significantly moving toward cloud-based systems, not only in terms of monitoring intrusion systems, but also with CCTV customers are demanding that the cost of securing their homes and business become less and less, as well as taking advantage of the lower maintenance requirement for hardware,” O’Callaghan says.
O’Callaghan bridles at the idea cloud is a component of IP-based electronic security applications of the future.
“The Risco cloud is here now and provides the major component of our service to customers, offering firmware updates, and connectivity for installers to their systems, allowing remote maintenance as well as the ability for end user to perform all manner of tasks remotely, including Live View, access doors, turn on lighting and heating,” he says.
According to O’Callaghan, there are a number of vital elements in any cloud-based electronic security solution.
“The most important element in a cloud based solution for electronic security is security, this goes without saying, and Risco has worked very hard at building its architecture to ensure the integrity of its network,” he says.
“Second comes scalability – the system needs to be able to support a continuing increase in systems being connected to the network, which just doesn't mean panels, but also the increase in the number of other connections as well, such as smartphones, computers, IP CCTV cameras, etc. Thirdly there’s ease of installation and use and Risco products have been designed with the installer in mind.”
When it comes to the greatest challenge for the future of cloud security applications in Australia, O’Callaghan maintains it’s about people.
“The greatest challenge would be building trust with the Australian consumer, and all the other challenges are a part of achieving this level of trust,” he says. “Meanwhile, the greatest strength of the Risco cloud, is providing accessibility to the end user for a whole range of services that were never available before, and now that is cost effective.”
According to Wai King Wong of Axis Communications, cloud is best defined by what it offers installers and end users.
“In the terms of security applications, cloud solutions enable a cost effective and trouble-free solution for end users that uses the latest hardware and software to deliver security via cloud,” explains Wong. “Axis does provide such solution, which is the Axis Video Hosting Solution (AVHS), and we believe that as networks become more widely available and more data centres are established in the country, takeup of this solution will grow.”
According to Wong, cloud will be a key component of IP-based electronic security applications in the future.
“It will be the future step to move into cloud-based solutions where systems can be in either private cloud (within a private network across a wide area network), or public cloud solution which relies on hosting companies,” says Wong. “One of the key benefits is the utilisation of hardware in a more efficient manner.”
What are the most important elements of any cloud-based electronic security solution – for instance, adequate layers of infrastructure, the user interface, lack of latency, encryption, other?
“With cloud-based solutions, the user experience is highly important,” says Wong. “Key requirements include low latency and highly secure transmission of information. We have a hybrid cloud solution where there is a secure connection from the camera to our cloud-based mediator server that communicates with the client and cameras. The solution is called Axis Camera Companion. The difference is that the recording is done locally in the SD card and the secure communication relies on the cloud-based mediator server.”
What is the greatest challenge for the future of cloud security applications in Australia – is it infrastructure? Is it bandwidth? Is it data security? Is it the security of mobile devices? Is it establishing a viable business model? Is it cost/economies of scale for the end user? Is it misapprehension in the market? Is it something else?
“The biggest challenge for cloud solutions in Australia will be infrastructure,” says Wong. “Australia is a large country and to provide high bandwidth everywhere is a big task. Therefore the NBN project is greatly needed in order for us to move to the next phase of cloud-based security solution.”
What do you see as being the greatest strengths of cloud? For instance, storage redundancy, redundancy of power, removal of the need to maintain network hardware and a secure network space on site, reduction in costs to power and cool network hardware, delivery of powerful remote capabilities to simple local security systems, the ability to allow remote management of systems, etc?
“The biggest strength of cloud-based solutions is its ability to harness the investment of hardware/software to reduce cost per user as each user shares the cost of hardware and software with a bigger pool of users,” says Wong. “That’s the key to cloud.”
There is, of course, another key to cloud. It's not a weather formation. It's a networked ecosystem than stretches from your security solution to a datacentre and onwards to authorised workstations and mobile devices. The variables associated with cloud – particularly latency – make it feel rather organic but there's mysterious about it. It's just computers on a globally accessible 'LAN'.♦
