3. Data Integration Architectures Flashcards
Modular Systems / Holistic Approach vs. Best of Breed Approach:
Modular Systems/Holistic Approach: This approach involves a central database with independent modules for each functional unit. It offers data consistency and extensibility but may suffer from low flexibility.
Best of Breed Approach: BoB involves selecting specialized software for each function, resulting in a system that is highly flexible and adapted to specific requirements. However, it may lead to functional incongruence and higher administrative efforts.
Data Exchange via Interfaces,
Involves the use of different schemata and formats for data exchange, necessitating standardization to overcome design autonomy
Communication Server:
Acts as middleware to manage interfaced systems. It simplifies mapping formats and protocols, increases manufacturer independence, and offers central maintenance but doesn’t completely resolve data integration issues.”
OLTP vs. OLAP:
OLTP (Online Transactional Processing): Focuses on administration and control, involving repetitive, structured operations primarily for update-in-place semantics.
OLAP (Online Analytical Processing): Geared towards analysis and decision support, allowing ad-hoc access for interactive data exploration and statistical analysis, primarily reading and aggregating data.
Integration Approaches:
Virtual Integration: No central data pool; queries are translated into subqueries answered by component systems. It offers up-to-date data but may have slow response times.
Materialized Integration: Involves central database replication and redundant data storage. Data transformation and integration occur offline, and timeliness depends on ETL process frequency.
Materialized Integration: Data Warehouse:
OLAP Cubes: Used in data warehouses for multidimensional analysis, providing comprehensive query options and supporting data mining methods.
Dimensions, Facts, Star Schema: Data warehouses often use a star schema, which includes dimensions (descriptive attributes) and facts (measurable quantities). This schema supports efficient querying and analysis.
key components of a modular system in HIS architecture?
central database, independent modules for functional units, and decouples modules from the database management system (DBMS) through a client-server concept.
adv, disadv of a modular system in HIS?
+++ data consistency, extensibility, and the avoidance of media discontinuities.
- Disadvantages include low flexibility and possibly insufficient functionality.
Best of Breed (BoB) approach in HIS and disadvantages?
involves selecting the best software for each purpose, resulting in a system with high flexibility and adaptation to requirements.
Disadvantages include functional incongruence, high risk of inconsistency, and high administrative effort.
types of autonomy in distributed data management
design autonomy, association autonomy, access autonomy, communication autonomy, and execution autonomy.
consequence of autonomy in terms of system heterogeneity?
autonomy leads to heterogeneity in runtime environments, communication protocols, data representation, data models, and semantics.
main challenge in integrating heterogeneous systems?
managing the different vendors, architectural approaches, and the resulting fragmentation
integration requirements in HIS?
overcoming heterogeneity, managing fragmentation, and ensuring information integration
methods of data exchange in HIS?
interfaces, communication servers, and repositories.
Key data exchange standards in HIS?
DICOM (Digital Imaging and Communications in Medicine) and HL7 (Health Level 7).
