Build and Deployment Process for Web Applications

[article]
Summary:
This paper describes practices that have led to a sound and reliable build and deployment process at Hewlett-Packard. Two teams of engineers, later joined by a third, responsible for developing e-service components to build a Web application, chose to use open source development tools/utilities in the "Evolutionary Software Development Lifecycle" environment.

Introduction
Software build process is an activity that compiles the source code, links it with other components and produces binaries that are ready to be deployed for production. Without a successful build, software cannot go into production. The build process brings together essential component of a software product and must be in place as early as possible in the software lifecycle. In his article on “Best Practices” [1], Steve McConnell has emphasized the importance of early and frequent builds. McConnell sites minimum integration risks, reduced low quality risks, and easier defect diagnosis as some of the benefits of early and frequent builds. Michel Cusumano and Richard W. Selby [2] have described the daily builds as the sync pulse of a project.

A Web application is made up of one or more e-services and thus requires a complex build and deployment operation. The factors such as linking with the correct version of an e-service, maintaining correct configurations of development, build and production servers make this operation complex. Once in production, any versioning requires an immediate and accurate build and deployment activity placing further demands on the process.

Challenges of Open Source Environment
Developing Web applications in the open source environment poses unique requirements on a build process. Streamlining the build process is complex due to numerous open source software tools and utilities, which are often developed and released separately. Generally, versioning and support of these tools and utilities seem to be a major problem. A typical utility is made available in the following builds:

  • Release—ready for prime time
  • Milestone—Unknown degree of quality
  • Nightly—very unstable

Only the release version is intended for production. These versions can often be downloaded at no cost. It encourages the developers to download the latest version for production. This behavior encourages unintended use of milestone and nightly build versions. Depending upon the nature of the developers, a development environment often ends up with multiple versions. Once a developer has acquired the version she wants, it often takes a fair amount of efforts to configure the utility and put it to it’s intended use. The documentation is limited and the only way to obtain some support is by using FAQ which has a long waiting period and does not match the development environment. This support will most likely not be synchronized with the needs of our development cycle, and will not be specific or in the context of our environment.

Although not specific to the open source environment, the other challenge faced by the development team is to build an application on multiple servers. The task becomes increasingly difficult if the servers have different configurations. The build and deployment process has to work on multiple platforms. Sanjay Mahapatra[3] has alluded to the benefits of platform independent builds.

Lastly, the development environment varies greatly. There are several IDEs (integrated development environments) such as Jbuilder, Visualcafe, and Netbeans, that make their way into the development environment. Although they do not cause major problems, they normally increase the complexity of the build and deployment process.

Project Information and Development Methodology
For the sake of confidentiality the project and the e-services names have been disguised. The mission of project XYZ was to build a Web application to support internet printing using three e-Services, LServe, MServe, and Nserve. These e-Services provided the underlining functionality. Each e-Service was owned by a development team and each team had a “build miester”. The Mserve team was also building another Web application. LServe had a dependency on e-Service Mserve and so MServe played a leading role for LServe.

The application was based on J2EE and supported by

Pages

About the author

Bhushan Gupta's picture Bhushan Gupta

Bhushan Gupta has nineteen years of experience in software engineering, nine of which have been in the software industry. Currently a Process Engineering Architect at Hewlett-Packard, he joined the company as a software quality engineer in 1997 where he was responsible for identifying key process areas to reduce rework. Based upon Hewlett-Packard's quality parameters, he has developed quality goals, quality plans, and software validation strategies for several products. In his current position he has promoted the iterative lifecycle, contributed to software validation planning and execution processes, carried out project retrospectives, and software process improvements. He has worked on use cases and their adoption for development of requirements and test cases. From 1995-97 he worked as a Systems Analyst at Consolidated Freightways where he contributed to the design and development of a Windows based logistic management system. During his ten years tenure (1985-1995) at the Oregon Institute of Technology Bhushan Gupta developed, planned, and taught numerous software engineering courses. He served as the curriculum coordinator for five years and resigned from his duties as an associate professor.

StickyMinds is one of the growing communities of the TechWell network.

Featuring fresh, insightful stories, TechWell.com is the place to go for what is happening in software development and delivery.  Join the conversation now!

Upcoming Events

Nov 09
Nov 09
Apr 13
May 03