Data Management Systems versus Data Warehousing – What Does Really Matter? Conclusion
The comparison between data warehouses and data management systems as central integration platforms demonstrates that data integration is the primary domain of warehouses, whereas ‘near-realtime’ processing and reconciliation are data management system strengths. In reverse, landscapes centred around data management systems are challenged with the integration of data from other systems and data warehouse approaches have to face the increasing requirements for ‘near-realtime’ processing and programmatic reconciliation.
Comprehensive solutions are not available off-the-shelf for either scenario, but require substantial customisation or custom development. Even if it is not appropriate to recommend one or the other type of clinical system landscape, there appears to be a clear trend within the data warehousing community in favour of a solution based on warehousing methodology. This is substantiated by expanding concepts for existing data warehouse platforms with ‘near-realtime’ processing, data reconciliation facilities and to enable data management processes within data warehouse environments. One can speculate about a long-term strategy to develop data warehouses that will fully support all data management tasks.
The perspective of data integration was taken in this article to compare the two types of landscapes. One important conclusion going beyond this comparison is that the same principles of semantic clarification and data modelling have to be applied to any system and environment, irrespective of the underlying architecture, in order to achieve data integration. As a consequence, data modelling capabilities, representation of metadata and management of
1. Mauron C, Methods for Standardization in Data Management Environments: Pharma iQ Clinical Data Management and Standardization Conference, 8–9 June 2010, London.
metadata have to be considered as important factors when assessing data handling systems and environments. These capabilities in particular are becoming increasingly important as the value of clinical data beyond submissions is being recognised by the industry. For example, upcoming methodologies such as business intelligence require flexible access to data integrated across diverse data sources. Of course, semantic integration is at best accomplished in an environment that does not impose substantial restrictions on data models; thus, the semantic perspective also favours the trend towards data warehouse-based approaches. n
Norbert Fritz is Development Leader of the Business Intelligence Warehouse in Product Development at F Hoffmann-La Roche Ltd in Basel. His role focuses on the development of an integrated data warehouse environment for the company’s information management group. After working in neuroscience at the University of Munich, he joined the Department of Neuropharmacology of Roche Switzerland. In 1994, he moved to pharma development at Roche where he
covered different positions. Dr Fritz studied biology, mathematics and informatics, has an MSc in biology and a PhD in zoology.
Charlotte Mauron is Manager, Data Modelling and Acquisition in Biometrics at F Hoffmann-La Roche Ltd in Basel. She has worked in several domains of clinical data management, database design and tools development for clinical research in multiple pharmaceutical companies. Since 1994 at Roche, she has held different roles focusing on global data modelling environment and metadata management. She is also involved in internal and external standardisation
initiatives. Ms Mauron studied biology and applied sciences and has an MSc in human physiology and chemistry and a post-graduate degree in computer science.
2. Brackett MB, The Data Warehouse Challenge: Taming Data Chaos, John Wiley & Sons, 1996.
3. Fritz N, Clinical data and the e-clinical landscape: A need for integration, Pharma Focus Asia,
2009;12:48–55.
4. Kimball R, Ross M, The Data Warehouse Toolkit. The Complete Guide to Dimensional Modeling, John Wiley & Sons, 2002.
DRUG DEVELOPMENT
47
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68