Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Model-Driven Software Migration: Process Model, Tool Support, and Application

Model-Driven Software Migration: Process Model, Tool Support, and Application

Ratings: (0)|Views: 59|Likes:
Published by Raaj Khanna
Legacy to SOA migration
Legacy to SOA migration

More info:

Categories:Types, Research
Published by: Raaj Khanna on Feb 03, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

07/27/2013

pdf

text

original

 
154
Copyright © 2013, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Chapter 7
INTRODUCTION
Most commercially built information systemsare based on traditional technologies preventingthem from unfolding their full potential in futuresoftware development. Service-Oriented Archi-tectures (SOA) (Arsanjani, et al., 2008; Gold,Knight, Mohan, & Munro, 2004) provide a modernand promising approach to increase flexibility insoftware adaptation, maintenance and evolution by referring to the underlying business processesto be supported by the software systems. Neces-sary functionality is specified by services which
Andreas Fuhr
University of Koblenz-Landau, Germany
Andreas Winter
Carl von Ossietzky University, Germany
Uwe Erdmenger
 Pro et Con GmbH, Germany
Tassilo Horn
University of Koblenz-Landau, Germany
Uwe Kaiser
 Pro et Con GmbH, Germany
Volker Riediger
University of Koblenz-Landau, Germany
Model-Driven SoftwareMigration:
Process Model, Tool Support,and Application
ABSTRACT
 Established software systems usually represent important assets, which are worth preserving in new software structures, to combine already proven functionality with the benets of new technologies. TheSOAMIG project is aimed at developing an adaptable migration process model with an accompanying tool support based on model-driven technologies. This process model, which combines reverse and  forward engineering techniques, was applied in two different case studies on migrating a monolithic software system to service-oriented architecture and to a transformation-based language migration from COBOL to Java.
Werner Teppe
 Amadeus Germany GmbH, Germany
DOI: 10.4018/978-1-4666-2488-7.ch007
 
155
Model-Driven Software Migration
are implemented by loosely coupled componentsthat can easily be rearranged to fulfill changinguser needs.Current software systems, already establishedand time-proven, are not implemented in a service-oriented manner. These legacy systems usually provide software functionality by monolithic anddeeply interwoven modules. Without significantreengineering, these systems will not benefit fromService-oriented Architectures. Referring to thestaged model for the software life cycle (Rajlich& Bennett, 2000), appropriate techniques arerequested to keep them in evolution instead of  passing them to service or replacing them.Software migration, i.e., transferring existingsoftware systems to a new environment, provideselaborated methods and techniques allowingalready established software systems to benefitfrom the advantages of new technologies. Mi-gration does not change the functionality of theoriginal systems; it only changes their embeddingin appropriate environments. Although migrationactivities use the same technologies as reengineer-ing activities, migration and reengineering haveto be distinguished from each other. Whereasreengineering focuses on improving softwarequality, migration only focuses on conservinglegacy systems in new environments (Sneed, Wolf,& Heilmann, 2010). Migrating legacy systems toSOA, as depicted in this section, allows the reuseof already established and proven software assetsand their integration with newly created services,including their further evolution (Fuhr, Horn,Riediger, & Winter, 2011). Orchestrating migratedand newly realized service components to sup- port new business processes is done in additional projects; the migration only enables the reuse of legacy functionality in a new service-orientedembedding. Migration to a new environment isnot viewed as activity to improve code quality.This allows clearly separating migration, i.e.,transformation, tasks from reengineering, i.e.,quality improvement, tasks. Nevertheless, migra-tion projects should be framed by reengineering projects to improve the legacy’s quality prior to(pre-renovation) and the migrated system’s qualityafter migration (post-renovation).Migrating to SOA addresses architecturemigration, which describes essentially rework-ing the software structure. Here, software assetshave to be identified and abstracted to servicessupporting business processes. Furthermore, codehas to be reorganized in new service componentsimplementing the extracted services. These mi-grations deal with various aspects of softwaresystems: data and databases, code and users andsystem interfaces are affected during migration.Occasionally, migration to SOA comes withlanguage migration where—next to structural ad-aptations—programming languages like COBOLto Java are changed as well.Migrating software systems requires a delib-erated process taking into account the differentaspects of migration projects. This process has to be supported by an adequate migration tool sup- port. Empirical studies in the METAMORPHOS project (de Lucia & Tortora, 2010) impressivelydemonstrated the increase of productivity in mi-gration projects by using tailored migration toolsuites instead of traditional software developmentframeworks. SOAMIG´s objective was to definesuch a process and to present and evaluate a cor-responding tool chain. Here, the main purpose wasto show its applicability. Discussion, assessment,and improvement of migration and reengineeringefforts were not part of the SOAMIG project. Theimprovement of maintenance efforts based on ap- propriate methodology and tooling is describedelsewhere (see e.g. Borchers, 1997; de Lucia &Tortora, 2010).Current process models in software develop-ment are primarily focused on the initial develop-ment of software systems. More modern, itera-tive and agile approaches like Unified Process(Kruchten, 2003), Extreme Programming (Beck & Andres, 2004) or SCRUM (Schwaber, 2004)view maintenance and migration activities only asadditional iterations. These process models do not
 
156
Model-Driven Software Migration
consider reverse engineering activities like com- prehending programs, identifying and evaluatingreusable software assets and legacy conversion.Hence, there is also no significant tool supportgiven for migration activities. Furthermore, migra-tion projects have to deal with legacy and targetarchitecture, including their interdependencies.Particularly in SOA migration, these architectureshave to be aligned to business processes (Fuhr,et al., 2011; de Lucia, Francese, Scanniello, &Tortora, 2008). In the same vein, approaches for developing Service-oriented Architectures onlyfocus on its initial creation. Here, e.g., SOMA(Arsanjani, et al., 2008) presents an expandableapproach, which was partially completed to in-clude migration support by using model-driventechniques (Fuhr, et al., 2011). Khadka et al. (2011)merge SOA development processes and extendthem by feasibility analysis techniques. The V-Model XT (IABG, 2009) provides a specialized project type merging enhancement and migration.Here, legacy analyses and the definition of amigration strategy are considered, but migrationin V-Model XT only focuses on data migration.Specialized migration approaches only addresscertain facets of migration projects. Accordingly,tool support is—if presented—also restricted tothese facets. The Chicken Little approach (Brodie& Stonebraker, 1995) defines an incremental pro-cess for migration of software systems in contrastto Cold-Turkey approaches, where migrations are performed and delivered at one go. Chicken Littlefocuses on long-lasting migration projects and is based on running legacy and migrated systems in parallel by using appropriate gateways and wrap- pers, which might lead to additional complexity of migration projects. Fleurey et al. (2007) presenta general model-driven migration approach notfocused on migration to SOA. Like the V-ModelXT, the Butterfly approach (Wu, et al., 1997)focuses on data migration. The reengineering fac-tory approach by Jens Borchers (1997) targets thecoordination of outsourced migration projects andSMART (Lewis & Smith, 2008) assists in plan-ning migration projects. A conceptual framework to SOA migration similar to Winter and Ziemann(2007) is being developed in the SAPIENSA proj-ect (Razavian & Lago, 2010; Razavian, Nguyen,Lago, & van den Heuvel, 2010). Both view legacyand target systems on code, architecture, and business process/requirements level and pursuea model-based migration strategy. Oldevik et al.(2011) present a model-driven SOA migrationapproach lacking a specialized tool support for reverse engineering activities. The ReMIP (Refer-ence Migration Process) (Gipp & Winter, 2007;Sneed, et al., 2010) introduces a general andadaptable process model for software migration.The migration process presented in Section 2substantiates the ReMIP, complements appropriatetool support and shows its application in industrialcase studies (see Sections 3-5).The SOAMIG project aimed at providing ageneral and adaptable migration process modelincluding a customizable tool suite supporting themigration process. In particular, SOAMIG sup- ports architecture migration to Service-orientedArchitectures and language migration. In contrastto wrapping-based migrations which hamper further evolution of the migrated systems, SO-AMIG considers transformation-based languagemigration. Transformation-based language mi-gration converts legacy systems completely intoa new (language) environment, without keepingthe legacy in its original form and adding morecomplexity by further different languages. TheSOAMIG project consortium consisted of indus-trial and academic technology providers, researchinstitutes and industrial application providers,which allowed practical validation of the SOAMIG process in two different migrations (Zillmann, etal., 2011). The SOAMIG project was partly funded by the German Federal Ministry of Education andResearch program for small and medium sizeenterprises (Grant no. 01IS09017A-D).The SOAMIG approach follows a model-basedmanner (Winter & Ziemann, 2007) with respectto the SEI horseshoe model (Kazman, Woods,

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->