The schema design
for data warehouses is usually very different from the fully normalized design best
suited for OLTP data stores. And data warehouses can support ad hoc queries that,
because of their complexity and the amount of data accessed, can place significant
loads on a system with only a handful of requests.
Reporting and query functions are part of an OLTP system, but the scope and frequency
are typically more controlled than in a data warehouse environment. For
example, a banking OLTP system will include queries for customer status and
account balances, but not multiyear transaction patterns.
The OLTP system typically provides forms that allow well-targeted queries that are
executed efficiently and don??™t consume undue resources. However, hard and fast
rules??”for example, that OLTP systems don??™t include extensive query facilities??”
don??™t necessarily hold true. The I/O performed by most OLTP systems tends to be
approximately 70??“80 percent read and 20??“30 percent write. Most transactions
involve the querying of data, such as product codes, customer names, account balances,
inventory levels, and so on.
Pages:
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456