Diagnostic Key DS6-029–042
DS6-029 — A · Hierarchical
A mandatory tree relationship where each child has one parent is hierarchical. Relations/tables, unrestricted graphs, and key-value schema-on-read are other organizations.
Source: p. 180 · Confusion: hierarchical vs relational.
DS6-030 — B · Relational
Schema-on-write, relations/tuples, row-oriented operational access, and SQL are the source clues. Schema-on-read/graphs, time-only data, and triples point elsewhere.
Source: pp. 180–181 · Confusion: relational vs non-relational.
DS6-031 — C · Non-relational / NoSQL
Chapter 6 allows schema-on-read, varied structures, and softer consistency postures. “Not Only SQL” does not mean no SQL, and NoSQL is not guaranteed stronger than ACID.
Source: p. 181 · Confusion: schema-on-read vs schema-on-write.
DS6-032 — D · Column-oriented
OLAP scanning few columns across millions of rows is the column-orientation clue. Row orientation better fits whole-row transaction access.
Source: pp. 181–182 · Confusion: OLTP row vs OLAP column.
DS6-033 — A · Valid vs transaction time
Valid time = real-world truth interval; transaction time = interval the database recorded/considered the fact true. Bi-temporal explicitly tracks both rather than making them synonyms.
Source: p. 181 · Confusion: valid vs transaction time.
DS6-034 — B · Spatial database
Geometric objects/functions—distance, area, overlap, buffer—identify spatial use.
Source: p. 182 · Confusion: spatial vs specialized databases.
DS6-035 — C · Key-value vs triplestore
Key-value retrieves values via keys; triplestore stores subject–predicate–object relationships. They are distinct database patterns, not backup/replication methods.
Source: pp. 183–184 · Confusion: key-value vs triplestore.
DS6-036 — D · Archiving
Rarely used but still retrievable later → archive. Purge is irreversible; sharding partitions; abstraction changes access interface.
Source: pp. 184–185 · Confusion: archive vs purge.
DS6-037 — A · CDC
Only changed rows identified from timestamps/logs is Change Data Capture. Full refresh moves all data; purge removes; federation presents autonomous sources.
Source: pp. 185–186 · Confusion: CDC vs full refresh / replication.
DS6-038 — B · Purging
Purge permanently removes obsolete/unneeded data so it cannot be recovered. Archive preserves retrieval; retention defines duration; replication maintains copies.
Source: p. 186 · Confusion: purge vs archive vs retention.
DS6-039 — C · Passive replication
One primary is updated and its resulting state propagates to secondaries. Active replication has replicas recreating state from operations; timestamp detection is CDC; combined autonomous view is federation.
Source: pp. 186–187 · Confusion: active vs passive replication.
DS6-040 — D · Log shipping
Applying transaction logs to a secondary every 15 minutes is the deciding clue. Mirroring is relatively immediate; sharding partitions; “active federation” is not the pattern.
Source: pp. 186–187 · Confusion: mirroring vs log shipping.
DS6-041 — A · Resiliency
Resiliency is tolerance of error conditions while continuing expected operation. Purge speed, number of replicas, and retention duration measure other things.
Source: pp. 187–188 · Confusion: resiliency vs recovery.
DS6-042 — B · Retention policy
Seven-year availability followed by deletion is a retention requirement that drives archive/purge specifications. Abstraction, schema ownership, and query optimization do not set lifecycle duration.
Source: p. 188 · Confusion: retention vs archive/purge.