Data Warehouse Research Project:
Towards Zero-Latency Data Warehousing
|
|
|
[Vision] [News] [People] [QTool] [Downloads] [Links]
Towards Zero-Latency Data Warehousing:
closing the feedback loop on continuous information integration and automated information delivery.
enable analytical decisions integrated into operational processes with minimized latency between cause and effect of business decisions.
Requirements:
Support for various data freshness requirements based on service levels in order to minimize propagation delays for particular data
Continuous data integration in order to enable near real-time capturing and loading of data
Timeliness of data and systematically dealing with late-arriving data in order to represent the history in the data warehouse as accurately as possible
Active decision engine based on the ECA (event-condition-action) paradigm:
events: data streams, conditions: (business) rules, actions: recommendations, notifications, decisions
High availability, scalability, and performance
Active Warehouse Workshop (16.07.2002), slides available on request.
New version of QTool (V. 2.4., 16.07.2002)
The QTool (Version 2.4 - 16.07.2002) is a research prototype including following features:
- source feeder
read transaction messages from a file and put them into a queue (MSMQ), arrival rate is adjustable- MSMQ administration tool
comfortable administration for MSMQ queues (much more administration features than the Windows-integrated computer management snap-in)- integrated TPump scheduler
generating scripts in order to run TPump and BTEQ jobs to continuously integrate data into a Teradata data warehouseFor further questions contact: bruckner@ifs.tuwien.ac.at, roland.braito@aon.at
Slides "Active Warehouse - Proof of Concept in a W2K / WinXP Environment", Teradata User Group Meeting, 22.05.2002, Vienna, Austria.
Institute of Software Technology & Interactive Systems - Department of Information & Software Engineering
Letzte Änderung: 10.10.2002