?‚? The treatment of security issues in ETL; source data and data in transit
are security risks (Friedman, 2002a).
??? Stream.ETL: Streams are sequences of data, continuously flowing from a data
source with the particular characteristic that, due to their volume, each tuple
is available only for a limited time window for querying. Stream examples
would involve stock rates extracted from the Web, packets going through a
router, clickstreams from a Web site, and so forth. Stream ETL is an ETL
process involving the possible filtering, value conversion, and transformations
of this incoming information in a relational format. Although, streams
cannot be stored, some patterns or snapshot aggregates of them can be stored
Data Warehouse Refreshment 2
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
for subsequent querying. As research challenges in this area, we can mention
the following issues:
?‚? The necessity of maintaining data in the DW as much ???online??? as we can,
but without adding an extra load to sources or DW.
?‚? The provision of correctness guarantees.
?‚? The necessity of cost models for the tuning of the incoming stream within
specified time window.
?‚? The audit of the incoming stream data for several constraints or business
rules, also with respect to stored data (e.
Pages:
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265