To efficiently handle organizational transactions, master data is used as context for transactional data. Master data is a critical component of organizational data and master data management techniques are used to handle this. Commercial MDM solutions are of high cost and designed for large organizations. Due to high costs, Small and Medium-size Enterprises (SMEs) are unable to adopt Commercial MDM solutions. As an alternative to commercial solutions and to improve the response time of cloud-based MDM solutions, this paper provides a fog-cloud-based architecture to handle the master data. Presented architecture provides distributed access and handles the issues associated with the cloud-based MDM solution of SME. iFogSim simulator has been used to validate the proposed architecture Performance of the presented architecture was evaluated and compared with the cloud-based solution using simulation.
For enterprise activities related to quality assessment, updating, and enhancement are dependent on master data. Master data is operational data of the organization which requires minimal updates. Changes in master data are very difficult to implement and to perform changes in master data, special provision is required. Master data is a kind of operational data that has high worth, characterizes center data that helps in a basic dynamic and taking care of business forms over the venture. Master data management (MDM) is a process used to handle the master data. MDM is responsible for the creation, updating, and deletion of the master data [8, 9,12, 14]. Master Data Management is used for customer relationship management, client integration, employee relationship management, quality management, and other management activities. Master data provides information about business perspectives. It gives the base to the transactional data. Master is always non-transactional data and it provide basic attributes of the business. It includes person, place, price, item, and other enterprisespecific data [11,13-15].
MDM is a process, which is used to provide atomicity for master data, enhance the quality of data and process flow in an organization. Due to cost constraints SME’s use simple MDM solutions based on either cloud architecture or a stand-alone approach. Cloud-based architecture generates high delay and faces network issues. In this study, we proposed a Fog-cloud architecture-based MDM solution. The proposed architecture uses fog computing approach to handle the response time and data access for MDM. According to this approach a copy of master data will be present in the fog layer that is near to the user. A complete description of this approach is given in third section. Section 2 provides a brief history of MDM. The fourth section provides results and discussion followed by the future scope and conclusion in the last section.
As per the studied literature, Siebel described the different categories of data used in organizations and defined the master data [14,15]. Many commercial MDM solutions were provided by big ventures like SAP, IBM, Oracle, Infosys, IBM, Google, Informatics, and TCS, etc. Apart from these big ventures, the research community also participated in this field. According to studied literature, it is observed that big ventures like IBM, SAP, etc. provided the solution of MDM for the large enterprises whereas after decade MDM solutions were also provided for the medium and small size enterprises using hybrid techniques, semantic frameworks, graph structures, and cloud-based solutions [2-6, 11, 14, 15]. MDM framework, maturity model, case studies were designed by the research community to understand the challenges, issues, and requirements associated with MDM [17,19-21]. A recent study highlighted the impact of COVID-19 on MDM [8]. To highlight the research journey in the field of master data, we provided an analysis of major publication from the year 1995 to till data in Figure 1. s100 n o i t a c i l b pu 50 f o t n u C 0 o 2016-Till date
Cloud based solutions are frequently used for SME having multiple operational sites. As per the studied literature, following are the key issues associated with SME based MDM solutions [1,7,9,13,18]: • • • •
Presence of high cost MDM solution Lack of unstructured solution for SME Presence of cloud based or stand-alone MDM solution for SME Delay in processing the data of SME due to limitations associated with cloud architecture for cloud-based MDM
For Small and Medium size enterprises, maintaining data integrity and consistency in existing MDM solutions is very expensive. Present standardize solutions for MDM are expensive for SME to adopt. Alternate to these solutions they use simple, cost-effective solutions. Most of these solutions are based on cloud computing approach and lack of strong internet may cause delay in data processing and slow system response. This paper focuses on the master data handling fog computing approach Ac-cording to this approach a copy of master data will be present in the fog layer that is near to the user. A complete description of this approach is given in the next section.
Cloud computing is frequently used for business applications and organizations. Cloud computing-based applications are facing issues like reduced speed, poor spectral performance, high latency, low connectivity, and security concerns [16]. SME’s create their own MDM solutions and these solutions are based on cloud or stand-alone system architecture. Cloud-based architecture generates high delay and faces network issues. The delay in processing is due to internet connection and traffic at the cloud channel [10,13,16,18].
Master data is very frequently used as reference data by users for daily operations but requires very fewer changes so considering this feature a Fog-cloud based architecture for the MDM process is proposed in this paper. 1. Fog layer linked with cloud layer as observer, User ask for the data using User Interface 2. User system generate data query as per the user request 3. Query is executed at fog layer for Master data 4. Fog Layer respond to user 5. As per the update massage from cloud layer, fog layer holds the transaction and update the master data 6. Now fog layer process user queries using updated data
To validate Fog-cloud-based architecture for the MDM, the iFogSim simulator has been used. Data from food enterprise has been used for this experiment. Two operational sites have been considered in this simulation. Food items, their prize, location of joint, and owner details have been considered as master data. To check the performance, 100 queries have been executed on MDM and without the fog layer. Execution Time-based comparison of Fog-cloud-based architecture for the MDM with cloud-based approach using food joint data is shown in Figure 4. Accuracy-based comparison of Fog-cloud-based architecture for the MDM with cloud-based approach using food joint data 5. As per the Figure 4, Fog-cloud-based architecture for the MDM has less access time in comparison to the cloud-based approach. Figure 6 shows the effect fog architecture on response time with respect to no of query parameter. Though Fog-cloud based architecture for the MDM approach requires additional space for maintaining multiple copies of the master data. According to Figure 5, Fog-cloud based architecture for the MDM and cloud-based MDM have the same accuracy. This study checks query result accuracy based on 100 queries. These bulk queries were implemented using the SQL loader concept. One major challenge with Fog-cloud based architecture for the MDM is maintenance and update handling at the fog layer. An additional mechanism is required to make the consistency at both cloud and fog layers.
1 2 3 4 5 6 7 8 Without Fog Layer 1 2 3 4 5 6 7 8 9
10
Market and economic operations depend heavily on master data. The academic group and the IT industry have produced several MDM technologies and frameworks. Due to high-cost SMEs cannot afford commercial MDM solutions and many cloud-based solutions were presented by the research community. These cloud-based solutions faced the issue of low response time and high processing delay. This paper provided fog-cloud-based architecture for MDM. iFogSim simulator has been used to validate the proposed architecture. Data of food joint with 100 queries have been used to validate Fog-cloud-based architecture. The proposed architecture provided the same accuracy and better performance than normal cloudbased architecture. Additional concurrency algorithm is required to maintain data consistency at both cloud and fog layers. One major challenge associated with the proposed architecture is to handle user queries while performing the update at the fog layer. In future, this approach can be implemented with consistency and concurrency control protocol. Security at the fog layer can also be added in future research.
[1] Al-Qerem, Ahmad, Mohammad Alauthman, Ammar Almomani, and B. B. Gupta. "IoT transaction processing through cooperative concurrency control on fog–cloud computing environment." Soft Computing 24, no. 8 (2020): 5695-5711. [2] Akhmetshin, E., Ilyina, I., Kulibanova, V., Teor, T.: Special aspects of master data- based integrated management of region reputation in modern it environment. In: IOP Conference Series: Materials Science and Engineering. vol. 497, p. 012022. IOP Publishing (2019) [3] Athanasiadou, I.: Evaluating the maturity of companies in supplier master data management: The design of a maturity model (2019) [4] Fernando, L.K., Haddela, P.S.: Hybrid framework for master data management. In: 2017 Seventeenth International Conference on Advances in ICT for Emerging Regions. pp. 1–7. IEEE (2017) [5] Ganesan, B., Parkala, S., Singh, N.R., Bhatia, S., Mishra, G., Pasha, M.A., Patel, H., Naganna, S.: Link prediction using graph neural networks for master data management. arXiv preprint arXiv:2003.04732 (2020) [6] Haneem, F., Kama, N., Taskin, N., Pauleen, D., Bakar, N.A.A.: Determinants of master data management adoption by local government organizations: An empirical study. International Journal of Information Management 45, 25–43 (2019) [7] Kaddour, Sidi Mohammed, and Mohamed Lehsaini. "Electricity Consumption Data Analysis Using Various Outlier Detection Methods." International Journal of Software Science and Computational Intelligence (IJSSCI) 13, no. 3 (2021): 12-27. [8] Khillari, S.: Impact of coronavirus on master data management market— growth, trends and forecast report, 2026 (2020) [9] Kokemuller, J., Weisbecker, A.: Master data management: Products and research. In: ICIQ. pp.
8–18. Citeseer (2009) [10] Mani, Nag, Melody Moh, and Teng-Sheng Moh. "Defending deep learning models against adversarial attacks." International Journal of Software Science and Computational Intelligence (IJSSCI) 13, no. 1 (2021): 72-89. [11] Murthy, K., Deshpande, P.M., Dey, A., Halasipuram, R., Mohania, M., Deepak, P., Reed, J., Schumacher, S.: Exploiting evidence from unstructured data to enhance master data management. Proceedings of the VLDB Endowment 5(12), 1862–1873 (2012) [12] Qodarsih, N., Yudhoatmojo, S.B., Hidayanto, A.N.: Master data management maturity assessment: A case study in the supreme court of the republic of Indonesia. In: 2018 6th International Conference on Cyber and IT Service Management (CITSM). pp. 1–7. IEEE (2018) [13] Siner, V.: Master data management in the EU. comparative analysis of access to base registries (2020) [14] Singh, S., Singh, J.: SSMDM: An approach of big data for semantically master data management. In: 2015 2nd International Conference on Computing for Sustainable Global Development (INDIACom). pp. 586–590. IEEE (2015) [15] Singh, S., Singh, J.: Management of SME’s semi structured data using semantic technique. In:
Applied Big Data Analytics in Operations Management, pp. 133-164. IGI Global (2017) [16] Singh, S., Singh, J.: Location driven edge assisted device and solutions for intelligent transportation. Fog, Edge, and Pervasive Computing in Intelligent IoT Driven Applications pp. 123–147 (2020) [17] Spruit, M., Pietzka, K.: Md3m: The master data management maturity model. Computers in
Human Behavior 51, 1068–1076 (2015) [18] Stergiou, Christos L., Konstantinos E. Psannis, and Brij B. Gupta. "IoT-based big data secure management in the fog over a 6G wireless network." IEEE Internet of Things Journal 8, no. 7 (2020): 5164-5171. [19] Subotic, D., Jovanovic, V., Poscic, P.: Data warehouse and master data management evolution-a meta-data-vault approach. Issues in Information Systems 15(2) (2014) [20] Zhao, C., Ren, L., Zhang, Z., Meng, Z.: Master data management for manufacturing big data: a method of evaluation for data network. World Wide Web 23(2), 1407– 1421 (2020) [21] Zuniga, D.V., Cruz, R.K., Ibanez, C.R., Dominguez, F., Moguerza, J.M.: Master data management maturity model for the microfinance sector in Peru. In: Proceedings of the 2nd International Conference on Information System and Data Mining. pp. 49–53 (2018)Wang, Xin, Tapani Ahonen, and Jari Nurmi. "Applying CDMA technique to network-on-chip." IEEE transactions on very large-scale integration (VLSI) systems 15.10 (2007): 1091-1100.