By Andreea Andrei, Marketing and Business Administration Executive at The Cloud Computing, SaaS and Security Awards

This article is part of an A to Z series by Cloud and SaaS Awards, continuing with P for Platform as a Service (PaaS) 

Particularly when it comes to business tools, technology is continually improving to make people’s work simpler. The growth in software and the global expansion of Internet connections gave rise to the concept of cloud computing, which is now widely used. Companies can use the cloud computing idea internally (private cloud), publicly (public cloud), or in a mixed way.

Platform as a Service (PaaS) is a term that refers to a cloud service through which the provider supplies the client with an environment for development, as well as the necessary tools for the development of new applications. Therefore, Platform as a Service is especially useful for developers and software companies who want to create and publish new programs or applications without having to personally take care of the necessary infrastructure.

The basic concept of Platform as a Service (PaaS)

The degree of resource centralization varies between different types of cloud services. Thus,  with the Platform as a Service (PaaS) approach, the customer has the option to deploy apps that they have purchased or generated themselves using the programming languages and tools that the provider supports.

The network, servers, operating systems, and storage components of the cloud infrastructure are not managed or under the control of the consumer, but you do have control over the installed apps and perhaps the hosting environment’s configuration. Google App Engine, Windows Azure, and Force by SalesForce are a few of the most significant platform services.

Developers or software firms are the typical clients of PaaS providers, whereas end users are served by the software companies’ SaaS offerings. Up to 2015, the PaaS industry already expanded at an annual rate of more than 30%.

The public cloud is primarily driven by economic benefits. The shift from a paradigm of capital investments in hardware and software to an operational expenditure infrastructure model and platform is how it manifests. Utilizing SaaS and PaaS models is strongly encouraged by these developments and the extra advantages of the public cloud approach.

The extent to which these platforms are used depends on how items are licensed and priced in relation to installation costs because the public cloud is entirely an on-going expense for users. The amount of disparity will affect how quickly adoption occurs.

Case studies: Amazon, Google, Oracle, and IBM

PaaS issues to be addressed

Speed enhancements and cost savings are the main forces behind the adoption of services. These services rely on the transmission of data processed in the datacenters of the PaaS providers and then returned to the user. Latency, the delay imposed on by the transmission capacity, becomes a critical concern when a business migrates their programs to the cloud.

As users become accustomed to high yields, their tolerance for latency falls when Internet services are given at a greater level. In order to limit the amount of information that has to be sent across nodes and to effectively use their services, cloud service providers need to approach the market with positioning strategies for local server farms. This will rely on how the infrastructure for sustaining the increased data flow develops.

Composition of Costs in PaaS Model

When a company opts for the PaaS model, they are solely liable for the variable operating costs related to the use of the resources from the previous month. This implies that these expenses may, in the majority of cases, be reimbursed by profits made based on the modality and payment terms of the clients. Nonetheless, these are the main costs to be considered in a PaaS model:

  • Compute (Windows Azure Instances);
  • Storage;
  • Storage Transactions;
  • 1GB Databases;
  • 10GB Databases;
  • AppFabric Service Bus Connections;
  • AppFabric Access Control Transactions;
  • Bandwidth (inbound);
  • Bandwidth (outbound);
  • IT Administration and Support;
  • Configuration and Commissioning.

What are the advantages of PaaS?

At the operational level, it is evident that the PaaS model has several advantages over the conventional models, such as:

  • Capacity Planning: IT services are bought on the basis of demand, on actual requirements and workload. Whenever improved performance is required or a new service is introduced,  this may be assigned fast and efficiently. Lower TCO (Total Cost of Ownership) is driven by dynamic scalability of capacity on demand. The associated contracts for yearly support and maintenance, as well as the server hardware (including failover and redundancy), software (including system operating, virtualization, management backup systems, and management security) can all be deleted.
  • Storage: The PaaS paradigm incorporates on-demand, persistent, highly scalable storage infrastructure. Storage equipment, software purchases, and contracts relating to support and maintenance can all be dropped. This involves getting rid of duplicate multimedia, multimedia services, and data backup software libraries (external media storage).
  • Servers and Database Software: This architecture is quite affordable, even when considering database licensing, data, and storage. Purchases of database server hardware, software, and server licenses (including failover and redundancy), operational system licenses, licenses for management systems, support contracts, and maintenance can all be eliminated.
  • Network: Best management practices are also provided, along with redundant network infrastructure.
    It is possible to get rid of management services, annual support and maintenance contracts, and network gear (including failover and redundancy).

In conclusion, anytime a project is dimensioned and is considered which servers are to be used as hosts, disputes may occur between various business divisions. However, from the advantages, break out of the costs and issues that may arise, every person in the firm should be content with the PaaS model.

The finance manager of the project would be relieved to not have to make an upfront server investment and to know that the company would only be paying for the consumption that takes place in the datacenter. On the other hand, the technical team would feel more secure knowing that the technology supporting it will never become saturated and that, in the event of rapid growth, you won’t have to extend their working hours to three daily shifts to set up new servers or migrate the applications to others, older ones.

Depending on the method chosen, the infrastructure required for implementation may even be unnecessary if all resources are outsourced.