Cloud Service Layer
The concept of delivering software as a service rather than a product you buy and install locally on your computer has been with us a long time now. In recent years, for many types of application, such as sales-force automation, web servers, games, email and other office applications, it has become a major route to market.
By utilising broadband access, data and server farms located at strategic hubs on the internet, it is possible to use a software application, access, edit and save data, view and review content, communicate with colleagues etc – without knowing or caring where the application is hosted or where the data is saved.
The concept has tremendous value to the consumer and business-user alike: no more concerns about server and network capacity, back-up, about failing disk drives, about software installation, configuration, product license keys and upgrades. Cloud delivers great economies of scale, tremendous elasticity in terms of spend commitment and the computing resources you can turn on and off as your business requires.
Secure networking plus data/server centres are core skill areas of Communication Service Providers (CSP). They are also trusted and valued suppliers to many Small-Medium Enterprises (SME) and corporate customers. As a result, many CSPs are leading the charge and providing the infrastructure to enable the provision of Enterprise IT applications to their customers ‘in the Cloud’.
But there is another angle to ‘Cloud’ that should be considered. Mobile telecom networks consist of a radio access network (RAN), a core network and the signalling layer. The RAN consists of fixed assets on the ground, providing the coverage and basic network access. Although comprised of interconnected computing resources, they need to be on the ground, near the base-stations, towers and switching centres. They are not candidates for ‘the Cloud’. The core network and the service layer which build on the RAN are a different story.
Telecom Service Layer in the CloudShow More…
The telecom signalling layer – the Service Layer – can be conceived as a layer above the connectivity of the core network which delivers the telecom services we all use and depend on every day. Indeed, the Service Layer is the main battleground for CSPs to deliver value to their customers beyond the basic commodity of the bit-pipe. The Service Layer consists of networked real-time computing resources with some specific and demanding operational requirements. If these specific operational requirements can be addressed adequately, then the Service Layer is a strong candidate to be delivered ‘in the Cloud’.
Traditional telecoms is built on the principle of “it is always available, it always works”. In contrast to the banking sector, for example – another service area which also has some specific and demanding operational requirements – where the systems are typically taken off-line for maintenance and upgrades at 3am on Monday mornings for maintenance and upgrades. The telecom network is always on, always available.
Specifically, the 'carrier-class' attributes of the Service Layer are: very low latency communication between communicating computers, control over the quality of service, security and traditionally at least, “five nines” availability (i.e. the system is available 99.999% of the time, measured through the whole year). These exacting requirements need special consideration when thinking of moving the Service Layer to ‘the Cloud’.
Given that customers value the carrier-class services delivered by CSPs, it is likely that any concerted move by CSPs of the Telecom Service Layer to ‘the Cloud’ will need to continue to deliver these key service differentiators to preserve their customers’ loyalty.
To host the Service Layer in ‘the Cloud’, a low latency, guaranteed Quality of Service (QoS) connection between the core network and the server farm delivering the computing is required. The signalling network connection also needs to have low-latency and a managed, guaranteed QoS, which in practise is likely to mean a private ‘Telecom Cloud’ rather than a best-efforts internet cloud.
Regardless of whether the Service Layer is delivered in a private Telecom Cloud or a best-efforts public Cloud, the computing equipment used – the Service Delivery Platform (SDP) – must be deployable and function in one of the performant virtualisation environments to reap the scalability/ elasticity rewards of virtualisation. Then, by pointing the core telecom network to the virtualised, SDP in ‘the Cloud’, CSPs can deliver carrier-class cloud-based services and network intelligence to their customers.
Already, major mobile operator groups are using OpenCloud products to deliver group-level, pan-country and pan-CSP telecom services in a Telecom Cloud. The prospect of CSPs reaping the scalability and price-elasticity benefits of Cloud and providing carrier-class telecom services on a truly global basis is surely imminent.Show Less…