Statistics gathering can be done in parallel.
Since Oracle Database 10g, statistics gathering is automatic and
populates the Automatic Workload Repository. For example, the
SQL Access Advisor leverages this information when making tuning
recommendations.
Summary Tables
Data within dimensions is usually hierarchical in nature (e.g., in the time dimension,
day rolls up to week, which rolls up to month, which rolls up to quarter, which rolls
up to year). If the query is simply looking for data summarized at a monthly level,
why should it have to sort through more detailed daily and weekly data? Instead, it
can simply view data at or above that level of the hierarchy. Formerly, data
warehousing performance consultants designed these types of summary tables??”
including multiple levels of precalculated summarization. For example, all the time
periods listed in Figure 10-2 can be calculated on the fly using different groupings of
days. However, to speed queries based on a different time series, a data warehouse
can have values precalculated for weeks and months and stored in summary tables to
which queries can be redirected.
Pages:
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512