Many people, quite correctly, point to Salesforce.com—which launched in February of 2000—as the first example of cloud computing. In the years following, the term ‘cloud’ became so popular and was applied to so many products and service offerings, that it became almost meaningless. The result was cloud confusion in the business world and in the media.
In one stark example from 2008, Oracle CEO Larry Ellison described news stories about cloud as, ‘complete gibberish.’ Endeavouring to provide some clarity on cloud, the U.S. National Institute of Standards and Technology (NIST) worked for more than two years to put together ‘The NIST Definition of Cloud Computing,’ which defines the five essential cloud computing characteristics and several ‘as-a-service’ cloud computing models.
True cloud applications
In the security industry, we talk about ‘True Cloud.’ It’s a term coined by Dean Drako, Founder and Former President and CEO of Barracuda Networks, now Founder and CEO of Eagle Eye Networks, IC Manage and Drako Motors. It refers to software running in the cloud that has the five essential cloud computing characteristics.
It refers to software running in the cloud that has the five essential cloud computing characteristics
A true cloud system is built from the ground up and is designed to function in the cloud for optimum cost and performance, using the Software as a Service (SaaS) model as defined in the NIST document. Despite the efforts of NIST and others, false cloud marketing persists. A CIO magazine article coined the term ‘cloudwashing’ to describe ‘the practice of taking legacy software and running it on a cloud instance, while calling it ‘true cloud.’
Physical security industry
So, it should not be surprising that within the physical security industry, many vendors have been loosely using the term ‘cloud’ to market any kind of software deployment in a public cloud data centre (such as AWS or Azure) as a ‘cloud offering,’ implying some level of conformance to the NIST cloud computing definition and its SaaS model, even though there isn’t.
Virtualisation and Orchestration are secrets to popularity of cloud popularity. In early 2022, DataArt reported that 94% of all business workloads are now processed on the cloud (75% is SaaS), and that half of all enterprise businesses expect to become cloud-native in 2022 (i.e., all or nearly all applications will be true cloud). A close look at the cloud computing essential characteristics reveals how and why cloud computing is becoming so popular.
Operating system software
The essential characteristics of cloud are established by using two software techniques
The essential characteristics of cloud are established by using two software techniques: virtualisation and orchestration. Virtualisation is a technique of using software to make virtual computing environments that act like physical hardware, but they are really software representations of the capabilities of underlying hardware.
This can be used, for example, to split a single physical server into multiple ‘virtual’ servers, making it appear as though each virtual server's operating system software is running on its own dedicated computer hardware and allowing each operating system to be rebooted and run independently. It can also be used in the opposite way, making the memory and processing resources of multiple computers work together as if they were a single large computer.
Individual physical server
The main difference between cloud virtualisation and traditional virtualisation is that very large, virtualised pools of computing resources are automatically managed by cloud virtualisation software, whereas traditional virtualisation relies on manual processes to adjust the sharing of computing resources on each individual physical server.
Orchestration is the automated configuration and coordination of data centre computer systems
Orchestration is the automated configuration, management, and coordination of data centre computer systems, applications, and services – including cloud computing services. Orchestration is used to automate complex IT tasks and workflows by simplifying the management of large-scale computing environments. This is what makes self-service possible for cloud-based systems.
The five essentials
These are the five essential characteristics that make high-performing scalable cloud systems and applications possible:
- Resource pooling, as described above, is the most fundamental cloud characteristic. A set of virtualised computing resources are shared by many subscribers whose resource usage is isolated from that of other subscribers, which is referred to as multi-tenancy.
- On-demand self-service allows security integrators or end-users to expand their subscriptions when they want to change the number of cameras, change a camera’s recorded video resolution, or update the number of days for a camera’s video retention. These selections are made in an application dashboard and don’t require human intervention from the cloud application provider.
- Broad network access means that all user application functionality is available from anywhere an internet connection is available, analogue or network cameras can be connected via cloud-managed appliances at a facility, and cameras made for connecting directly to the cloud data centre can be placed practically anywhere.
- Rapid elasticity allows computing resources to be increased or decreased instantly, for example, to maintain sufficient video retention for outdoor cameras, which has traditionally been a challenge even for motion-based recording, in weeks when rain, wind or other activity spikes up the percentage of time motion recording is activated.
- Measured service meters the resources provided to ensure that each subscriber only uses and pays for the resources allotted by subscription.
Video management systems
That’s why vendors of all types are eager to market their products and services as cloud
The combination of these cloud characteristics means that video management systems are always provisioned correctly, eliminating concerns about system sizing and hitting CPU processing or storage capacity ceilings, and trying to prevent over-provisioning, which raises both fixed costs and variable operational costs in non-true-cloud systems.
More than two decades after the first cloud computing systems were launched, the benefits of the cloud are well understood. That’s why vendors of all types are eager to market their products and services as the cloud. However, it’s important to look beyond marketing. Pooling, rapid elasticity, and on-demand self-service are not easy to deploy securely and globally. It takes years of cloud computing engineering experience and a considerable amount of investment to get it right.
Will your system provide all of the benefits of a true system to both end-user and integrator? It’s worth taking the time to ask your vendor how the system conforms to the five essential characteristics of a true cloud system.