Native Apps At The Client & Cloud

Srinivasan Sundara Rajan

Subscribe to Srinivasan Sundara Rajan: eMailAlertsEmail Alerts
Get Srinivasan Sundara Rajan: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


Related Topics: Cloud Computing, Agile Software Development, SaaS Journal, Java Developer Magazine, Microsoft Developer, IT as a Service, Platform as a Service

Article

Software Development as a Service (DevaaS)

Agile and cloud - a winning combination

Runtime Benefits of Application Systems in the Cloud
With every major IT and telecom vendor announcing support for cloud every other day, the cloud is here to stay and no questions about it. However, most of the benefits are at runtime or when the IT applications are executed, for example:

  • We have options to consume applications from public clouds (SaaS)
  • We have options to utilize platforms of choice (PaaS)
  • We have production environments provisioned dynamically (IaaS)
  • We have environments scaled dynamically based on the usage patterns (elasticity and dynamic scaling)
  • We have metering and billing options (pay per use)
  • We share systems (multi-tenancy)

However, there aren't any case studies that show the value proposition of the cloud during the analysis, design, development and testing of IT applications.

IT Budget Allocations Waterfall vs Agile SDLC Models
Before the IT systems hit production and get the benefits of the runtime elasticity of the cloud platform, they go through a long and costly cycle of operations to get the systems to meet the needs of the business both functionally and non-functionally.

This cycle of process is generally referred to as SDLC (Software Development Life Cycle).

The IT team generally collaborates with business to obtain the budget allocation for building such systems. There are multiple ways that software systems go through their life cycle before they gets deployed to production:

Waterfall: The Waterfall Method is comprised of a series of very definite phases, each one run intended to be started sequentially only after the last has been completed, with one or more tangible deliverables produced at the end of each phase.

Essentially, it starts with a heavy, documented, requirements-planning-and-sign-off's phase that outlines all the requirements for the project, followed by sequential phases of design, coding, test-casing, optional documentation, verification (alpha-testing), validation (beta-testing), and finally deployment/release.

Agile: Agile methods break tasks into small increments with minimal planning, and do not directly involve long-term planning. Iterations are short time frames that typically last from one to four weeks. Each iteration involves a team working through a full software development cycle including planning, requirements analysis, design, coding, unit testing and acceptance testing when a working product is shown to the stake holders. This minimizes overall risk and allows the project to adapt to changes quickly. Stakeholders produce documentation as required.

Estimating The Project Size  And Budget Allocation  Waterfall Vs Agile

Waterfall

Agile

Requirements are frozen  and estimation cycles are long and detailed. Possible to predict the budget and effort.

Requirements evolve during the project and accurate estimates  and budgeting is not possible.

It is possible to identify, define, schedule,

and order all the detailed activities.

Near the beginning, it is not possible.

Adaptive steps driven by build-feedback

cycles are required.

Cloud Suitability for Waterfall SDLC

For Waterfall projects, the project resources such as:

  • Software needs
  • Hardware needs
  • Seating needs
  • Duration of the infrastructure rentals
  • Number of licensing needs

Can be predicted fairly accurately. While much of the advantages of the cloud are seen as an ‘Unpredictable' increase or decrease in needs. This theory leads us to think that adopting a Cloud platform for a Waterfall project may not produce much benefits in terms of budgetary and cost control.

Agile SDLC And Cloud

  • In agile estimating is iterative. Estimates change as more information emerges.
  • In a complex environment with frequent changes that cannot be anticipated, estimation must be done in an incremental and adaptive way.
  • Customers will have the option to scale up and scale down the
    • Number of requirements
    • Complexity of Requirements

Per iteration

The following alignment between Agile SDLC and Cloud will dramatically reduce the budgets allocated with the software development life cycle.

Software Licenses    =>   PaaS Offerings

Server, Infrastructure Needs   =>   IaaS Offerings

Productivity Tools  =>   SaaS Offerings

With a minimal team of key people involving the prime stake holders every thing else can be dynamically provisioned to support the Agile SDLC, as shown below.

The Missing Link - ‘Development as a Service'
The above diagram indicates that most of the unpredictable cost factors of the Agile SDLC (Software, Licenses, Tools, Platform and Servers) can be dynamically scaled using Cloud.

But what is missing are the resources, how an Agile organization can dynamically provision human resources to support the increase or decrease in the needs.

It brings the interesting concept of (Software) Development As A Service, so that the software development Components can be provisioned as a service, using the Cloud-based model.

The following diagram gives an outline of (Software) Development as a Service. This concept is not new, several IT Services organization have adopted Software Factories such that the if desired input is given for a software development task the same can be produced and delivered. The same concept can be extended to Cloud, in a catalogue and self service based provisioning and hence organizations adopting agile development can provision the software development needs on-demand basis.

Summary
As we have seen, the Agile SDLC method mitigates much of the risks of the  large water fall software development projects. This method also brings the  software development costs under control. With the adoption of  Cloud the cost of  Agile SDLC will go further down.

The  new thinking on DevaaS (Development As A Service), if realized can transform the way the software development can be thought out and executed.

More Stories By Srinivasan Sundara Rajan

Highly passionate about utilizing Digital Technologies to enable next generation enterprise. Believes in enterprise transformation through the Natives (Cloud Native & Mobile Native).