ods layer:
Data sources and modeling: The source data for each business system, the business model is consistent with the physical model;
services: providing data to other logical layer;
data ETL process described: the traffic data extraction landing into a text file, a reload ods layer to the data warehouse, not clean conversion.
dwd layer:
data sources and modeling: Data from ods layer, is the fact that DW detail layer, consistent data model ods layer;
service areas: to provide detailed data for the various business topics edw;
data ETL Process Description: According ods increment merge data to generate a full amount of data, the conversion is not washed, the total amount to preserve the original data.
dw layer:
data sources and modeling: Data from dwd layer, is the fact that DW layer, using dimensional modeling, star schema, this layer can be subdivided into dwb and dws;
service areas: to provide details of each topic for the EDS business ;
data ETL process description: lightly cleaned from dwd layer, conversion, generating dw layer summary data such as character combined, email, document number, date, phone number converting combined; summary by polymerizing the respective dimensions.
dm layer :
Data sources and modeling: Data from dw layer, using dimensional modeling, star schema;
service areas: data mining, custom query application marketplace;
data ETL process described: coarse-grained polymeric layer from dw summary, the facts according to business needs widening to form a wide table.
st layer:
Data sources and modeling: Data from dw layer, using dimensional modeling, star schema;
services: the front end of the report shows, subject analysis, KPI reports;
data ETL Process Description: coarse-grained aggregate rollup layer from dw such as year, month, season, day of the dimensions of some aggregate fact data generation business needs.