Building Business Agility: Public, private and hybrid cloud computing models for utilities

Public, private and hybrid cloud computing models for utilities

Clouds are data centres or multiple data centres comprising computation and storage resources connected by a network. They are smart, automated and adaptive. Applications can be deployed far more quickly without customer provisioning boxes and, once deployed, these applications can dynamically scale up on demand. Issues like congestion failure can be resolved automatically. The cloud allows businesses to get the resources they need without having the hardware on-premises. This means businesses can deploy services wherever and whenever they want. Cloud computing can reduce costs by allowing hardware to be consolidated and managed in fewer locations. It provides the scalability to support fluctuating workloads, and it can empower collaboration among remote teams and locations. To put it simply, clouds are more efficient and cost effective than traditional data centres. To achieve a scalable, demand-based smart grid IT infrastructure, cloud computing solutions and services must be incorporated.

Cloud deployment models

There can be different types of clouds including public, private and hybrid. The deployment model depends on the business needs and technical requirements. In general, users of private clouds enjoy higher levels of security and customisation to fulfil their regulatory needs and unique circumstances. On the other hand, users of public clouds enjoy relatively lower costs and on-demand scalability. Hybrid clouds bring a combination of these two benefits, with manageable shortcomings.

Public cloud: In the public cloud, infrastructure is made available to the general public or a large industry group and it is owned by an organisation selling cloud services on a commercial basis. The cloud service provider rents it to its mini customers and tenants, who generally pay only for services they actually use. Public cloud services require little upfront cost and can be deployed quickly. Public clouds also allow offloading of management, and hence, are the best option for hosting everyday functions such as emails, customer relationship management and human resources.

Private cloud: Private clouds have only one tenant, customised to specific needs. Such a cloud is operated solely for an organisation. It may be managed by the organisation itself or a third party that may exist on or off premises. Most businesses are moving to private clouds for running core business apps to provide complete comparative advantage like research, manufacturing and supply chain management. It requires a larger upfront cost and ongoing management, but it can deliver potential cost savings over the long term.

Hybrid cloud: Hybrid clouds are a combination of both public and private clouds, so as to share applications and data between them as needed. The clouds have the ability to allow data and/or applications to be moved from one cloud to another through their interfaces. One gets benefits of both and can have an additional benefit of creating innovative apps on demand. One can deploy it on private cloud and move to public cloud when demand spikes. This gives business the flexibility to run applications in a way that maximises cost savings and use of resources while meeting requirements for scalability and control. These benefits make the hybrid cloud the top cloud strategy among enterprises.

Overall, companies using hybrid clouds can still have higher security for mission-critical workloads and leverage low-cost public clouds for less sensitive workloads. The key feature of a hybrid cloud is the compatibility between the public and private clouds, which could be effectively managed.

Cloud service models

There are many types of cloud service models for utilities to choose from. The three main models that utilities can opt for, keeping in mind budgets and resources, are software-as-a-service (SaaS), platform-as-a-service (PaaS), function-as-a-service (FaaS) and infrastructure-as-a-service (IaaS).

In SaaS, the end-user application is delivered as a service. Users do not need to manage anything except for their data. Licences are purchased on a subscription basis and services are delivered immediately. Platform and infrastructure are abstracted and can be deployed and managed with less effort. Businesses use this for everyday operations. However, relying on internet access to function, SaaS applications tend to be slower than client/server applications. In many cases, SaaS cloud-based applications have less functionality and features than their client/server counterparts. This disadvantage, however, may be void if your business only needs the features offered in the SaaS version to function.

PaaS is an application platform on to which custom applications and services can be deployed. It offers hardware resources plus the operating system and databases. PaaS lets business develop, run and manage applications without having to build and maintain the infrastructure. Applications can be built more inexpensively, although services need to be managed. However, it is difficult to migrate many of the services provided by one PaaS product to a competing product, thus making it hard to switch PaaS vendors. Downtime and additional expenses are likely to occur while switching from one PaaS provider to another.

With FaaS, users manage only functions and data while the cloud provider manages the application. This allows developers to get the functions they need without paying for services when code is not running.

In IaaS, physical infrastructure is abstracted to provide computing, storage and networking as a service, avoiding the expense and need for dedicated systems. While this frees businesses from having to own or maintain hardware, managing of operating systems, databases and applications is still required. IaaS offers the most control and flexibility of all service models and can be easily scaled up or down as needed. IaaS models are typically much more expensive than PaaS and SaaS models because they offer much more support to business than the other two cloud models.

Besides the above models, there can be additional derivative service models of cloud computing. For example, Communications as a Service (CaaS) is also a cloud service category in which the capability provided to the cloud service customer is real-time interaction and collaboration. Second, Computer as a Service (CompaaS) is a cloud service category in which the capabilities provided to the cloud service customer are the provision and use of processing resources needed to deploy and run software. Third, Data Storage as a Service (DSaaS) is a category in which the capability provided to the cloud service customer is the provision and use of data storage and related capabilities. Last, Network as a Service (NaaS) is another category in which the capability provided to the cloud service customer is transport connectivity and related network capabilities.

The way forward

The grid modernisation journey for utilities may vary depending on an organisation’s starting point and its aspirations for agility. Companies may seek to shift the bulk of their operations to hybrid or public cloud operating environments, move discrete parts of their application portfolio, or eliminate particular legacy infrastructure platforms. Some utilities may prefer building applications based on their skills and competitive context over buying them, while others are highly selective in their building strategies and focus on integrating third-party SaaS solutions. Additional factors include a company’s industry, level of maturity, tolerance for risk and organisational readiness for pursuing agility.

User requirements are continuously evolving, as are the requirements for interfaces, networking and storage. This means that a cloud, especially a public one, does not remain static and is also continuously evolving. There are several government regulatory compliance concerns with cloud computing. Going ahead, given that moving to the cloud involves a huge investment commitment, analysing and choosing the right cloud computing and service model becomes critical.

Nikita Gupta