Cloud Migration Research: A Systematic Review  

·        Overview

·        Research Questions

·        Research Method

·        Characterization Framework

·        Data

·        Researchers

·        External Evaluations

·        Published Papers

Overview

Context: By leveraging cloud services, organizations can deploy their software systems over a pool of software and hardware resources. Deployment of organizational software systems to cloud offers organizations with differentiating advantages such as elasticity to fluctuations, pay-per-use and efficient utilization of the resources. On the other hand, organizations heavily depend on their business critical software systems, which have been developed over lifetime of the organizations, also known as legacy applications, which usually deployed on-premises. Over the past few years, cloud computing has gained a significant momentum and organizations want their legacy software to be migrated towards cloud-based environments. In recent years, research on migrating legacy software towards cloud-based environments have also proliferated. However, there is no secondary study to review state-of-the-art to systematically investigate a collective impact of research on legacy to cloud migration.

Objective: In this research, we aim to identify, taxonomically classify and systematically compare the existing research focused on migration of legacy systems towards cloud-based software. The results of a systematic investigation facilitate knowledge transfer among research community in terms of potential, limitations and future dimensions of existing research. This research provides a historic overview, focusing on methods and techniques in legacy to cloud migration and reflects on areas of future research.

Method: We conducted a systematic literature review (SLR) of 23 – qualitatively selected – studies, published from 2010 to 2013. We classify and compare the selected studies based on a characterization framework that we also introduced in this paper.

Results: To this end, 23 studies are taxonomically classified and holistically compared using a characterization framework. The research synthesis resulted in a knowledge base of current research approaches, methods, techniques, best practices and experiences used in legacy to cloud migration. This systematic review also resulted in an identification of the research gaps as well as several directions for future research.

Implications: Our review reveals that cloud migration research is still in its early stages of maturity but this is improving. This review identifies the needs for a migration framework to help improving the maturity level of cloud migration. Although contributions regarding migration decision support are popular in the selected studies, crosscutting concerns and migration execution are not in the main attention. This study also shows a lack of tool support to automate and facilitate cloud migration tasks.

Research Questions

RQ1: What are the main practical motivations behind legacy migrations towards the cloud?

RQ2: What are the main migration tasks involved and available tool supports?

RQ3: What are the existing methods and techniques to enable migration of legacy software towards cloud-based environment?

RQ4: What are the existing research issues and what should be the future research agenda in legacy to cloud migration?

Research Method

Our study followed the principles of a quasi-systematic review complemented by a seminal guideline and also documented learnt lessons in [pdf]. However, we extend it in a way that it can accommodate the characterization scheme comprising 12 data items. The following figure gives an overview of the three-phased methodology we applied in the study.

                                                                                 Overview of the review process

Characterization Framework

CharacterizationFramework

Data - our strategy is to update the data repository for cloud migration every 6 months, so please send us a copy of your published paper if its topic is on "cloud migration".

Researchers

The systematic search, extraction and mapping were conducted by the following researchers:

Pooyan Jamshidi - Aakash Ahmad - Claus Pahl

IC4 - The Irish Centre for Cloud Computing and Commerce

Lero - The Irish Software Engineering Research Centre

School of Computing, Dublin City University, Ireland

Post publication data updates are done by the following researchers:

Reza Teimourzadegan - Hamed Jamshidi

External Evaluations

The study was externally evaluated by the following active researchers and domain experts:

Please submit your comments and newly published papers that we can include in our data repository to: Pooyan Jamshidi

Published Paper(s)

P. Jamshidi, A. Ahmad, C. Pahl, "Cloud Migration Research: A Systematic Review", IEEE Transactions on Cloud Computing, vol. 1, issue 2, pp. 142-157, 2013, DOI: 10.1109/TCC.2013.10.