Databases Reference
In-Depth Information
FIGURE 8.3
Workload categories.
●
Workarounds—current workarounds include:
●
Minimizing data movement when processing large data sets by incorporating ETL and ETL-
hybrid architectures.
●
Special network connectors between the ETL server and the data warehouse to bypass the
enterprise network and avoid clogging it.
●
Adding infrastructure to process data and move data faster.
●
Implementing SOA-type architectures to data processing.
●
Goal—the goal for creating an effective workload strategy at this layer is to minimize the amount
of work to be done by either the ETL or the data warehouse or, in a different approach, maximize
each processing or iteration cycle where the resources are effectively utilized.
Query classification
There are primarily four categories of workload that are generated today: wide/wide, wide/narrow,
narrow/wide, and narrow/narrow (
Figure 8.3
).
Wide/Wide
Wide/wide workloads are queries from analytics or multidimensional analysis, and can also be trig-
gered by ad-hoc users and return long sets of columns typically joining more than two tables in the
result set. The resulting complexity from these queries are impacted by the:
●
Length of the table
●
Distribution of the table across the storage
●
Data model
●
Data relationship
