Diagnostics — DWB11-018–034
DWB11-018 — B
Why B: shared governed Date/Product dimensions across marts are conformed dimensions, enabling cross-process comparison.
A: ODS is current operational integration.
C: Data Vault hub is a different historical modeling construct.
D: full comparison is CDC.
Source: p. 368. · Confusion: conformed vs local dimensions.
DWB11-019 — C
Why C: the bus matrix crosses business processes/fact areas with dimensions to plan conformance/increments.
A: not an ETL schedule.
B: not network topology.
D: not access control.
Source: p. 368. · Confusion: bus matrix vs schedule/topology/security.
DWB11-020 — D
Why D: Inmon centers integration in a normalized enterprise warehouse; Kimball in dimensional increments connected by conformed dimensions/bus. Both target integrated analytical data.
A/B/C: falsely deny history/integration/BI capabilities to one approach.
Source: pp. 365–369. · Confusion: Inmon vs Kimball.
DWB11-021 — A
Why A: staging is an intermediate preparation area for cleansing, integration, enrichment, standardization and transformation before analytical targets.
B: BI portal is consumption.
C: summaries are presentation/optimization.
D: durable history belongs in DW.
Source: p. 371. · Confusion: staging vs analytical storage.
DWB11-022 — B
Why B: the central/atomic warehouse persists integrated historical atomic data and latest loaded state for analytical use.
A: current-day transactions alone are operational.
C: summary-only violates atomic foundation.
D: rejects belong to controlled recycle/error areas.
Source: p. 371. · Confusion: central warehouse vs current/summarized/recycle stores.
DWB11-023 — C
Why C: few-minute refresh + recent history + integrated operational use are ODS clues.
A: historical DW alone mismatches latency/current focus.
B: quarterly mart mismatches freshness.
D: Metadata repository describes data rather than serving this operational dataset.
Source: p. 371. · Confusion: ODS vs DW/mart.
DWB11-024 — D
Why D: a Finance historical analytical subset is a data mart.
A: staging is preparation.
B/C: message bus/transaction log are integration/change evidence, not analytical presentation stores.
Source: p. 371. · Confusion: data mart vs staging/integration evidence.
DWB11-025 — A
Why A: ROLAP, MOLAP and HOLAP correspond to relational, multidimensional and hybrid OLAP families.
B: integration patterns.
C: modeling/physical choices, not the named OLAP families.
D: decision/BI horizons.
Source: p. 371. · Confusion: OLAP implementation families.
DWB11-026 — B
Why B: historical loading is baseline/backfill usually done once/few times; ongoing update repeats to keep current.
A/C/D: invent cadence/content equivalences unsupported by the source.
Source: pp. 371–372. · Confusion: historical load vs ongoing update.
DWB11-027 — C
Why C: Data Vault can retain normalized atomic history/keys, supporting traceability and rebuilding/adapting downstream dimensional marts.
A: DQ still matters.
B: ODS is current/near-current, different purpose.
D: presentation can still be downstream; users need not query raw source.
Source: p. 372. · Confusion: Data Vault vs ODS/raw.
DWB11-028 — D
Why D: reliable update timestamps directly support time-stamped delta; lack of delete evidence is a known limitation.
A: transaction-log CDC requires DB log evidence.
B: message delta requires published events.
C: full comparison is not necessary when good timestamp evidence exists unless other requirements force it.
Source: pp. 372–373. · Confusion: timestamp CDC vs other CDC.
DWB11-029 — A
Why A: explicit insert/update/delete change table is exactly log-table delta CDC.
B: timestamps are weaker than the provided explicit table.
C: trickle is delivery cadence, not change-table evidence.
D: OLAP refresh is downstream analytical processing.
Source: p. 373. · Confusion: log-table CDC vs timestamp/low-latency delivery.
DWB11-030 — B
Why B: transaction-log CDC reads DB transaction evidence, can capture deletes and avoids timestamp overlap.
A: timestamp lacks inherent delete evidence.
C: full comparison does not use log evidence and is heavier.
D: trickle describes source mini-batches.
Source: p. 373. · Confusion: transaction-log CDC vs timestamp/full/trickle.
DWB11-031 — C
Why C: source-side mini-batches triggered by time/count are trickle feed.
A: messaging accumulates/distributes through a bus.
B: streaming clue is continuous target-side queue/buffer.
D: not historical backfill.
Source: pp. 373–374. · Confusion: trickle vs messaging/streaming.
DWB11-032 — D
Why D: Chapter 11 distinguishes messaging and streaming by accumulation: bus vs target queue/buffer.
A/B/C: impose false constraints unrelated to the source distinction.
Source: pp. 373–374. · Confusion: messaging vs streaming.
DWB11-033 — A
Why A: Understand Requirements is the first of the six core activities because business goals/questions/KPIs define scope.
B/C/D: population, BI implementation and maintenance follow earlier design/development work.
Source: p. 374. · Confusion: activity sequence.
DWB11-034 — B
Why B: requirements start with business goals, processes, questions, categories, KPIs and calculations.
A: procurement-first is tool-driven.
C: summary-first violates atomic/requirements principles.
D: evolving use does not eliminate requirements discovery.
Source: pp. 374–375. · Confusion: business requirements vs source/tool-first.