Browse Prior Art Database

Universal, Exclusion Based, Declaratively Driven, Light Weight, Highly Scalable, Softcopy Artifact Analysis for Relationship and Architecture Understanding

IP.com Disclosure Number: IPCOM000178025D
Original Publication Date: 2009-Jan-13
Included in the Prior Art Database: 2009-Jan-13
Document File: 4 page(s) / 40K

Publishing Venue

IBM

Abstract

The massive volume of software artifacts in an enterprise became hard to mange and hard keep current. Understanding of relationships between those artifacts helps to classify and group those artifacts and can lead to turning them into assets rather than liabilities. Software creation is still a very human activity. The interfaces created have distinct characteristics, specific to the domain. Finding potential interaction elements in software based on the fact that they have been human created for this domain and filtering them out of language syntax and semantics can allow for language and environment agnostic methods of finding relationships between software artifacts.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 32% of the total text.

Page 1 of 4

Universal, Exclusion Based, Declaratively Driven, Light Weight, Highly Scalable, Softcopy Artifact Analysis for Relationship and Architecture Understanding

The massive volume of software artifacts in an enterprise became hard to mange and hard keep current. Understanding of relationships between those artifacts helps to classify and group those artifacts and can lead to turning them into assets rather than liabilities. In the era of WEB 2.0, Service Oriented Architecture, and Resource Oriented Architecture, massive software artifact generation and exchange is happening on an unprecedented scale. Identification and maintenance of relationships and dependencies across software artifacts of heterogenous types became a significant problem. It applies to programs,

program source, runtime artifacts, documentation,

WEB site content, resources delivered using WEB Services, and resources published using RESTful services, even e-mail and instant messaging.

New alternatives are necessary to identify impacts of a seemingly local change, understanding

and discovering software artifact architecture on a global enterprise scale. Discovering and tracking the relationships and dependencies across software artifacts requires human knowledge of the software systems, but the volume is beyond the human ability to analyze those artifacts and memorize and persist the results in usable form.

Tools such as IBM Rational Asset Analyzer (RAA) can scan and analyze programmatic artifact relationships from multiple types of repositories such as SCMs, database metadata. There are other tools also performing static analysis of software artifacts that compose executable elements. There are separate tools that understand complex relationships of architecture of WEB sites, WEB services etc. There are also tools that trace and record parameters of execution of software. Static analysis is required to write code that understands the complex syntax and semantics of specific artifact types. In case of dynamic analysis, the platform and middleware runtime details must be taken into consideration when creating software for instrumentation of data capturing.

There are keyword technology based analysis and search tools (including WEB search, like Google). The keyword based technologies index the content of software artifacts, but can generate integrated indexes for heterogenous artifacts without detailed understanding of artifact semantics. Those tools cannot and do not attempt to do analysis or understand the structure of

programmatic artifacts. Keyword based tooling must store all of the data and perform significant

processing in order to maximize performance.

In the enterprise, various types of softcopy artifacts exist in various physical locations and various business departments. There are artifacts related to information technology including

program sources, runtime environment configurations on distributed platforms, documenta...