Books for you

OLTP Versus Business Intelligence
Mixed workloads??”OLTP and reporting??”are the source of many performance challenges
and the topic of intense debate. The data warehousing industry had its genesis
in the realization that OLTP systems could not realistically provide the needed transaction
throughput while supporting the enormous amount of historical data and ad hoc
query workload that business analysts needed for things like multiyear trend analysis.
The issue isn??™t simply one of adequate machine horsepower; rather, it??™s the way data
is modeled, stored, and accessed, which is typically quite different. In OLTP, the
design centers on analyzing and automating business processes to provide consistent
performance for a well-known set of transactions and users. The workload revolves
around large numbers of short and well-defined transactions??”with a fairly significant
percentage of write transactions.
Business intelligence typically operates on larger data stores that frequently are
assembled from multiple data sources and contain long histories.


Pages:
431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455