Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»

Of course, there are other important design aspects of the data warehouse that will be discussed in later chapters.

Partitioning of Data

In the data warehouse environment, the question is not whether current detail data will be partitioned but how current detail data will be partitioned. Figure 2.19 illustrates partitioning.

The purpose of partitioning of current detail data is to break data up into small, manageable physical units. Why is this so important? The operations staff and the designer have more flexibility in managing small physical units of data than large ones.

Following are some of the tasks that cannot easily be performed when data resides in large physical units:

■    Restructuring

■    Indexing

■    Sequential scanning, if needed

■    Reorganization

■■ Recovery ■■ Monitoring

In short, one of the essences of the data warehouse is the flexible access of data. Large masses of data defeat much of the purpose of the data warehouse. Therefore, all current-detail data warehouse data will be partitioned.

Data is partitioned when data of a like structure is divided into more than one physical unit of data. In addition, any given unit of data belongs to one and only one partition.

partitioning of data

processing complex A

Скачать в pdf «Building the Data Warehouse»