Database Administration In A Changing Environment: Special Reference To DMFAS, DSM+ And Debt-Pro

By Paddy Turyamwijuka
May 2005
Meeting and maintaining an organisation’s information needs present needs and many challenges to the data management profession. It is the aim of this paper to provide solutions and guidelines that help a debt data manager meet these needs.
A database is an excellent mechanism for getting information into the hands of decision makers. However, it is only as good as the data that goes into it. Problems occur when we attempt to acquire and deliver this information.
DMFAS users in Uganda have found it difficult to maintain an accurate and consistent DMFAS database both on an account of lack of proper security procedures in some parts of the database architecture and well defined roles of a DMFAS data administrator.
Both data generators and data users are under economic pressures to drive down the cost of their respective services. This pressure forces data generators to take short cuts and circumvent the data quality process. This combination of factors is very dangerous and has led to an untold number of situations where the end users absolute confidence in DMFAS data is unwarranted. Ironically, because these short cuts can dramatically impact price, both generators and users are rewarded by receiving additional work. This vicious cycle has led to a proliferation of “data time bombs” where data goes on to be used in reports for what may be an inappropriate use.
Knowing an error has occurred is fairly useless unless one can identify its source and prevent a re-occurrence. The system should be able to use triggers to serve as routine tracking procedures. Triggers can help track down historical record of normal processes.
This paper proposes six solutions to the problem: Paisley’s transactional model, tracking database modifications, defining roles of a DMFAS data administrator, training, Workshops, and DMFAS user certification.
The above solutions if included in the database model design, address the problem by providing a practical means to measuring and ensuring better data quality. Factors instrumental in practising good data management are education, accountability, and management support. Agency staff, including managers, must understand these data management principles and must then be held accountable for their use.
The paper takes on a case study (DSM+ vis-à-vis Debt-Pro) where numerous inquiries concerning proper data management have been raised by agencies and several workshops organised by MEFMI and DRI on Debt Sustainability Analysis. While the applications mentioned above can be considered already well established and both designed for debt sustainability analysis, the paper investigates whether there are differences between DSM+ and Debt-Pro input and results, if any, whether such differences may be caused by differences in data management approaches.
This case study primarily seeks to identify areas that require change if the systems used for DSA are to interoperate. This paper describes the similarities and differences between the Debt-Pro and the DSM+ object model. The differences can be viewed as requirements for changes in one or the other in order to promote data interoperability. Many differences represent concepts present in one of the models that are absent in the other. Resolution of some of these differences will require the addition of missing concepts to the deficient model. Other differences are differences in details of how common concepts are modelled. Here, negotiated agreements on the presentation of such details will need to be reached.
Generating answers to this question is the objective of data administration, data management, data alignment, and data transformation, which can be defined as the building blocks of a role in the comparability process. The first three of these tasks can be standardized and used a general manner. Only the task of data transformation is system dependent.
Most of the differences identified originate from the DSM+ Debt-Pro Timing Conventions regarding debt service and disbursements, net present calculation and differences in modes of restructuring.
To resolve the differences the paper proposes that a team be set up with terms of reference that include completing an inventory and analysis for both systems. Their first species of concern should be the timing convention, discuss the issue at length and agree on a modality that is line with debt strategy simulation framework.