Smart Continuous Delivery Framework for Software Releases
2016
In today`s IT world, there is an increasing demand for quality software products that helps business grow, which can in turn help reduce the effort in time & money and also be highly dependable.The challenge of having product releases frequently is not quite an easy task as it sounds.To meet this challenge of producing reliable software products, the IT managers & leads need a team of dedicated developers, system programmers, testers along with a highly efficient process in place.Continuous Delivery (CD) is a software engineering approach in which teams keep producing valuable software in short cycles and ensure that the software can be reliably released at any time.CD is attracting increasing attention and recognition.The continuous delivery mechanism already has certain frameworks such as agile framework, Scaled Agile Framework (SAFe), Disciplined Agile Delivery (DAD), Composable Fault Tolerance Framework (CFT), Test Orchestration Framework & Large-scale Scrum Framework (LeSS).These existing frameworks do not have any approach to determine or predict the futuristic state of the continuous delivery pipeline.Detecting problems early in development require a new CD approach that speeds up testing & eventually successful releases.This research work will propose a new framework that can provide error & problem prediction analysis & determine a desired futuristic state in software release lifecycle.The research proposes a new continuous delivery framework with features of early defect recognition through machine learning and that is smart to take informed decisions.This new framework can help the adoption of CD as a practice across IT organizations helping in effective handling of scenarios when the existing frameworks fail.The intent is to generate and maintain a data store to help in decision making for continuous delivery life cycle.The data will be collected & stored & will be mined to be analyzed for success & failure points & concluding on reasons & factors towards the outcomes.The data store will constitute the backbone of the machine learning that will be mined or possible outcome scenarios hence assisting in early detection of problems & help with its resolution in the continuous delivery lifecycle.Extending the adoption of the continuous delivery framework across organizations requires cloud based deployments as most organizations depend on cloud services to host the services.There are a plethora of cloud enablers available & the framework aims to provide the cloud enablement feature to help in its increased adoption.