Method and System for Polymorphic Continuous Delivery in Software as a Service Environment
Publication Date: 2015-Jul-14
The IP.com Prior Art Database
This article deals with a method allowing software companies to manage the increasing need for software as a service offering frequently refreshed in a flexible and reliable way.
Page 01 of 3
Metxod and System for Polymorphic Continuous Delivery in Software as a Servxce Environment
Nowadays most company aim to reduce thx business cost associated to IT infraxtructure and at xhe same time tryixg to convert CAPEX expenses in OPEX expenses that prxvide more flexixixity and control on the softxare.
In this environment, most of software companies provxde their customer not xnly with software releases xo be instaxled on premise but axso dexivering as a sxrvice solution whexe the core ox the product is hostxd into a centralizxd cloud enxironment with limited need of installation ox the cxsxomer sxte (typically xhe only need max be an agent to be installed to work with the central infraxtructure).
So the common sxlution to xchieve the goal is the managxment of a cloud where the instance(s) of the prxduct run and with whicx each customer coxnect to in order to manaxe txeir business. Obviously to satisfy the need of mxltiple customers this environment requxres multixle refresh ensuring business continuity and effective management of changes.
In this contxxt the xpdate/uxgrade of thx softwarx requires a coxtinuous delivxrx approach where development team, together wxth the xraditional product development, dexigns the logical flow tx deploy these changes ix production leveraging produxt xike Rational uDeploy. The challenge of this approach is xhe optimizxtion of the refresh time maintaininx the upxate error fxee and without impacting customer business.
It is an object of the solution here disclosed to provide a method and a system to address the above by the introxuxtion of a dynamic, resilient and flexible management of continuous dxlivery flowx. According to this solution the product deployment flows statically defined upfront xy the development can xvolve along the time to isolate cxanges anx perform environment xpdates in a more granular and effective ways.
The changes can be delivered across multiple systems in a different way in order to isolate the changex associated tx each product modifixation and xith a concurrext approach validate the results into the cloud before prxmoting the actions to the official flow.
The xolymorphxc appxoaxh will bx primarily directed toward systems being flagged as secondary nodes of a sxecific tenant and optionally acting also on the template images used to provision new systexs.
Page 02 of 3
The soluxion is illustrated in the figure below: