RELIABLE SOFTWARE COST ESTIMATION TECHNIQUE WITH TECHNICAL DEBT MANAGEMENT USING ORCHESTRATION

Transcription

1 International Journal of Advanced Research in Engineering and Technology (IJARET) Volume 9, Issue 6, November - December 2018, pp , Article ID: IJARET_09_06_011 Available online at ISSN Print: and ISSN Online: IAEME Publication RELIABLE SOFTWARE COST ESTIMATION TECHNIQUE WITH TECHNICAL DEBT MANAGEMENT USING ORCHESTRATION K. Ravikumar Research Scholar, Dept. of CSE, Manonmanium Sundaranar University, Thirunelveli Tamilnadu, India Dr. G. Gunasekaran Principal, JNN Institute of Engineering, Chennai, Tamilnadu, India ABSTRACT Software industry has developed various kinds of software product which is consumed and used by the customer. The delivery of the product is more accurate, high quality with less expensive in nature. There is some obstacle exists in the software which leads the performance problem due cost estimation. The costs are estimated in various factors from initial stage to delivery stage. Technical debt is one of the main elements in the cost estimation which grows the overall cost to high. Technical debt consists of defect, design, testing and etc. There are some methods available for debt reduction they are refractor, training and so on. The existing technical debt management is based on the single software organization which maintains their own technique for handling debt. It leads the reliability problem because of lack in the various form of debt which exists in the software development in an organization. This problem is overcome by using the integrated technical debt management over various software organizations with mutual agreement. The main objective of the proposed method is to identify various debts available in the different organization. This will help the software companies to follow common standard and also reduce the cost in overall estimation. Key words: Cost Estimation, Defect Management, Risk Management, Software Metrics, Technical Debt. Cite this Article: K. Ravikumar and Dr. G. Gunasekaran, Reliable Software Cost Estimation Technique with Technical Debt Management Using Orchestration. International Journal of Advanced Research in Engineering and Technology, 9(6), 2018, pp INTRODUCTION Nowadays the software grows enormously which fulfil the customer needs because everything is automated in the real world situation. The software company has to take care of everything goes wrong. The risks occur in the various forms which fall from software analysis phase to implementation phase. The estimation of software cost plays vital role in the editor@iaeme.com

2 K. Ravikumar and Dr. G. Gunasekaran software development process. The cost leads the performance problem in the software maintenance due to excess cost available in the cumulative manner. The technical debt which is available in the software should minimize then only the cost reduction is carried out in the software process management. Self-Admitted Technical Debt (SATD) focuses on the file level files without isolation category. This problem is overcome using change level mode. This mode uses various features with three dimensional view of software identification such as diffusion, message and history [1]. Temporal based granularity level of the analysis related to technical debt with repayment mode. It also responsible for repayment if in case of any violations occurs in the software development process [2]. The technical debt of the Machine Learning system can be monitored with suitable test bed which is related to the big data prediction analysis problem [3]. Technical debt leads the severe problem in the development of any software with high cost which is addressed by using automated version of software by considering performance attributes such as size of the code, complexity and API based modularity and so on[4]. The moral based analysis of the technical debt has been carried out based on the elements such as aim, affective property. It calculates the negative impact during the investincation process [5]. The systematic way of managing the software debt with integrated solution is implemented in production of software with various demission of analysis. This is also concentrates the maturity models level based analysis with suitable investigation [6]. The software development process targets only the fast delivery of the products without any latency. It cannot achieve maximum reliability because of the internal technical debt in the software. The existing problem is overcome by conducting the web oriented analysis from different user with their feedback [7]. Self admit is a technical debt which is available in the code with inappropriate solution. This problem is overcome by using TEDIOUS (Technical Debt Identification System based on the machine learning method with various metrics such as code based, readable based and analysis based. This is analyzed based on the precision and recall accuracy measurement [8]. The technical debt are analysed and identified over the software code with the parameters like identification process, impact over the debt, presence of the debt. This problem is overcome by using optimal way of analyzing and updating process [9]. The negative technical debt degrades the performance in the software product development without considering the cost and debt related to the architecture. The life time of the software is enhanced by keeping the software with optimal risk with technical debt [10]. The rest of the paper is organized as follows, Section 2 represents the architecture of the proposed method, section 3 describes that the algorithm of the proposed method. Section 4 shows that the experimental evaluation and section 5 represents the conclusion and future work. 2. ARCHITECTURE OF THE PROPOSED METHOD Software industry has follows their own standards for developing and maintaining the software which is delivered with high quality product. There two dimensions of software product development they are profit and loss. The product leads the positive dimension which proves that the software is high quality otherwise not. The risk maintenance also done based on the standard which used and deployed at the industry level. The software cost also goes high because of technical debt exist in the production environment. The technical debt minimization needs the software quality goes high with least cost. Initially the software organization establishes the agreement among them because of eliminating the disputes occur in the software development and integration process. This support various component of the editor@iaeme.com

3 Reliable Software Cost Estimation Technique with Technical Debt Management Using Orchestration software development among different organization are integrated which generates the generic solution. This solution is common to all organization in which the technical debt is minimized. The teams of the various maturity levels are identified which assessed the capabilities of the software organization. The requirements are maintained in the requirement register which is collected from various organizations which analyze the risk. The technical debt of each and every organization is accessed. The accessed debt are analyzed in two dimensions for achieving minimized cost in the software they are risk based and defect based. The risk based approach analyzes and identify the risk in order to minimize the risk because it also includes the software cost. The defect based approach analyzed and accessed from the corresponding register. The proposed architecture of the technical debt reduction is shown in figure 1. The defect level is assessed and eliminates it in order to improve the performance by achieving the cost reduction. There no risk and defect then the friction are identified from the software product. The integrated technical debts are identified and total debts are assessed over different organization. The statistics about the overall integrated debt minimization is recorded for the better quality. Figure 1. The overall architecture of the proposed method

4 K. Ravikumar and Dr. G. Gunasekaran 3. ALGORITHM Algorithm Integration_Technical_Debt_Minimization Begin Establish the agreement over software organizations; Let T is a Team; L1 : for each ti ε Team do Identify the Team ti; Collect the attributes from requirement register; Access the Technical debt; End. Risks are identified as RI; Defects are accessed as DA; For each risk ε RI do Analyse the risk; Risk minimization process; End For each defect ε DA do Defect Registers access; Defect level analysis; End Integrated friction identification; Friction Analysis; Integrates Technical debt analysis as ITDA If (ITPA>Theshold) Goto L1; Else Minimized technical debt; End; End 4. EXPERIMENTAL EVALUATION The technical debt can be analyzed with various parameters such as mean, minimum and maximum threshold level and standard deviation. The number of samples also considered for performance improvement in the software development. The debt are classified based on the types namely defect based debt, design level debt, documentation based debt, algorithm and testing based, architectural level debt. These debts are considered as technical debt, so this is reduced by using various methods such as refractor and inspection etc. The time and cost is a primary element while handling technical debt. These will be analyzed based on the violations, severity level such as high, medium and low etc. The estimation carried out with.net based application, C application, Java application, C++ application and networking application in both JAVA platform and.net platform. The estimates related to the technical debt is assessed and analyzed by the customer. This will be improved by adjusting some of the parameter in the technical debt with reliability editor@iaeme.com

5 Reliable Software Cost Estimation Technique with Technical Debt Management Using Orchestration manner. The severity of various applications is classified in to different level namely high, low and medium. The level of.net, JAVA, C, C++ application is 100%, 50%, 25%, 10% respectively. The proposed integrated technical debt minimization in various levels is shown in Fig.2, Fig. 3, Fig. 4 and Fig.5 respectively. Fig.6 describes that the integrated software application debt management with time and cost dimensions over software development process. Figure 2. Defect Debt Comparison Fgure 3. Architectural Debt Comparison Figure 4. Design Debt Comparison editor@iaeme.com

6 K. Ravikumar and Dr. G. Gunasekaran Figure 5. Algorithm Debt Comparison Figure 6. Integrated Cost Estimation Comparison 5. CONCLUSION AND FUTURE WORK Technical debt management plays important role in the software development process. This debt includes the cost estimation process which leads the extra expensive. The debt is not considered in the particular phase which accumulates the cost to the final estimation which leads the cost estimation goes in wrong projection. The number of violations occur during the development any software also the causes for technical debt. This type of debt is managed properly then only the overall cost estimation keep under the company s control. There are various types of technical debt available in the software so it can be identified and eliminated for better cost control. The proposed method integrates various software organizations with generic standard with mutual agreement. The standard gives overall debt management process across the software development. There are two type of analysis are also carried out they are risk based and defect based. The risk based analysis is to assess the level of risk exists in the software where as the defect based analysis to identified and eliminated, so that the cost estimation goes in proper direction. The main aim of the proposed approach gives the minimized cost estimation in various dimensions with least cost operation. In future the technical debt is analyzed in multidimensional perspective with real time situations with maximum reliability editor@iaeme.com

7 Reliable Software Cost Estimation Technique with Technical Debt Management Using Orchestration REFERENCES [1] Meng Yan, Xin Xia, Emad Shihab, David Lo, Jianwei Yin, Xiaohu Yang,"Automating Change-level Self-admitted Technical Debt Determination", IEEE Transactions on Software Engineering, DOI: /TSE , [2] Georgios Digkas,Mircea Lungu, Paris Avgeriou, Alexander Chatzigeorgiou, Apostolos Ampatzoglou,"How do developers fix issues and pay back technical debt in the Apache ecosystem?", International Conference on Software Analysis, Evolution and Reengineering (SANER), DOI: /SANER ,2018,Pages: [3] Eric Breck,hanqing Cai,Eric Nielsen, Michael Salib, D. Sculle,"The ML test score: A rubric for ML production readiness and technical debt reduction", International Conference on Big Data (Big Data) 2017, DOI: /BigData ,2017. Pages: [4] Lorena Capitán ; Birgit Vogel-Heuser,"Metrics for software quality in automated production systems as an indicator for technical debt", IEEE Conference on Automation Science and Engineering (CASE), DOI: /COASE , 2017.Pages: [5] Hadi Ghanbari,Terese Besker, Antonio Martini,Jan Bosch,"Looking for Peace of Mind? Manage Your (Technical) Debt: An Exploratory Field Study",ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM), DOI: /ESEM ,Pages: [6] Narayan Ramasubbu, Chris F Kemerer,"Integrating Technical Debt Management and Software Quality Management Processes: A Normative Framework and Field Tests",IEEE Transactions on Software Engineering, DOI: /TSE , [7] Terese Besker, Antonio Martini, Jan Bosch,"The Pricey Bill of Technical Debt: When and by Whom will it be Paid?", IEEE International Conference on Software Maintenance and Evolution (ICSME), 2017,Pages: [8] Fiorella Zampetti, Cedric Noiseux, Giuliano Antoniol, Foutse Khomh,Massimiliano Di Penta,"Recommending when Design Technical Debt Should be Self-Admitted",IEEE International Conference on Software Maintenance and Evolution (ICSME) Pages: [9] Everton Da S. Maldonado, Rabe Abdalkareem, Emad Shihab,Alexander Serebrenik,"An Empirical Study on the Removal of Self-Admitted Technical Debt",IEEE International Conference on Software Maintenance and Evolution (ICSME), 2017, DOI: /ICSME ,Pages: [10] Terese Besker, Antonio Martini, Jan Bosch,"Impact of Architectural Technical Debt on Daily Software Development Work A Survey of Software Practitioners",43rd Euromicro Conference on Software Engineering and Advanced Applications (SEAA), 2017, Pages: editor@iaeme.com