35/35
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
Z
Numerics
3NF,
2.3
A
advantages
In-Memory Column Store,
2.5.1
aggregates,
5.2.1
,
11.7.5
computability check,
11.1.3.4
aggregation
In-Memory Column Store,
19.8.6
VECTOR GROUP BY,
19.8.6
ALTER MATERIALIZED VIEW statement,
5.3
altering dimensions,
9.6
amortization
calculating,
21.5.6
analytic functions,
21.3.5
concepts,
18.1
analytic processing
materialized views,
6.2
applications
decision support systems (DSS),
4.1.2
architecture
data warehouse,
1.4
attribute clustering,
12.1
adding to existing table,
12.2.4.1
advantages,
12.1.2
data dictionary views,
12.3
dropping,
12.2.4.3
guidelines,
12.1.1.4
methods,
12.1
modifying,
12.2.4.2
privileges,
12.2.1
types,
12.1.1
interleaved ordering,
12.1.1.2
linear ordering,
12.1.1.1
,
12.2.2
,
12.2.3
using hints for DML,
12.2.4.4
attribute-clustered tables,
12.1
attributes,
2.2
,
9.2
automatic big table caching
about,
2.6
B
bind variables
with query rewrite,
11.3.11
bitmap indexes,
4.1.1
nulls and,
4.1.2.1.2
on partitioned tables,
4.1.2.1.3
parallel query and DML,
4.1.2
,
4.1.2
bitmap join indexes,
4.1.2.2
B-tree indexes,
4.1.3
bitmap indexes versus,
4.1.2.1
build methods,
5.3.4
business intelligence,
1.1
queries,
22
business rules
violation of,
17.4.1
C
cardinality
degree of,
4.1.2.1
CASE expressions,
18.7.3
cell referencing,
21.2.4
columns
cardinality,
4.1.2.1
common joins,
11.1.3.1.1
common tasks
in data warehouses,
1.3
COMPLETE clause,
5.3.7
complete refresh,
7.1.1
complex queries
snowflake schemas,
2.4.3
composite
columns,
19.6
compression
See data segment compression,
5.3.3
concatenated groupings,
19.7
concatenated ROLLUP,
19.10.1
configuring
In-Memory Column Store,
2.5
constraints,
4.2
,
9.4
foreign key,
4.2.2.2
RELY,
4.2.2.3
states,
4.2.1
unique,
4.2.2.1
view,
4.2.2.7
,
11.3.6
with partitioning,
4.2.2.6
with query rewrite,
11.7.1
cost-based rewrite,
11.1.1
CREATE DIMENSION statement,
9.2
CREATE MATERIALIZED VIEW statement,
5.3
enabling query rewrite,
10.2
creating
zone maps,
13.2.2.2
with attribute clustering,
13.2.2.1
CUBE clause,
19.3
partial,
19.3.3
when to use,
19.3.1
cubes
hierarchical,
6.2.1
materialized views,
6.2.1
CUME_DIST function,
18.2.1.3
D
data
nonvolatile,
1.1
purging,
7.5.3
,
7.5.3
sufficiency check,
11.1.3.2
transformation,
17.3
transportation,
16.2.1
data compression,
4.4.1
See data segment compression,
5.3.3
data cubes
hierarchical,
19.7.1
data densification,
18.5
time series calculation,
18.6
with sparse data,
18.5.2
data marts,
1.4.3
data rules
violation of,
17.4.2
data segment compression,
3.2.1.2
materialized views,
5.3.3
partitioning,
3.2.1.2
data transformation
multistage,
17.1.1.1
pipelined,
17.1.1.2
data warehouse,
5.1
architectures,
1.4
dimension tables,
5.1.6
fact tables,
5.1.6
physical design,
3.1
refresh tips,
7.1.10
data warehouses
common tasks,
1.3
key characteristics of,
1.1.1
database
staging,
5.1
date folding
with query rewrite,
11.3.5
DBMS_ADVISOR
TUNE_MVIEW procedure,
5.3.5
DBMS_ERROR package,
17.4.2.2
DBMS_MVIEW package,
7.1.6
,
7.1.6
EXPLAIN_MVIEW procedure,
5.8.1
EXPLAIN_REWRITE procedure,
11.6.2
REFRESH procedure,
7.1
,
7.1.7
REFRESH_ALL_MVIEWS procedure,
7.1
REFRESH_DEPENDENT procedure,
7.1
DBMS_STATS package,
11.1.1
DBMS_SYNC_REFRESH package,
8
decision support systems (DSS)
bitmap indexes,
4.1.2
degree of cardinality,
4.1.2.1
DENSE_RANK function,
18.2.1.1
densification
data,
18.5
design
logical,
3.1
physical,
3.1
dimension levels
skipping,
9.2
dimension tables,
5.1.6
normalized,
9.2.3
dimensions,
3.2.6
,
9.1
,
9.4
altering,
9.6
analyzing,
19.1.1
creating,
9.2
definition,
9.1
dimension tables,
5.1.6
dropping,
9.7
hierarchies,
3.2.6.1
multiple,
19.1.1
skipping levels,
9.2
validating,
9.5
with query rewrite,
11.7.2
drilling down,
9.1
hierarchies,
9.1
DROP MATERIALIZED VIEW statement,
5.3
prebuilt tables,
5.5
dropping
dimensions,
9.7
materialized views,
5.7
zone maps,
13.2.4
E
ELT,
14.1
ENFORCED mode,
10.2.3
entity,
2.2
error logging,
17.4
table,
17.4.2.2
errors
handling,
17.4
ETL. See extraction, transformation, and loading (ETL),
14.1
examples
zone maps
join pruning,
13.4.2.2
partition and table scan pruning,
13.4.2.1
EXCHANGE PARTITION statement,
4.2.2.6
execution plans
star transformations,
4.5.2.2
EXPAND_GSET_TO_UNION hint,
11.3.8.2
,
11.7.10.4
EXPLAIN PLAN statement,
11.6.1
star transformations,
4.5.2.2
EXPLAIN_REWRITE procedure,
11.6.2
exporting
EXP utility,
15.3.1.3
expression matching
with query rewrite,
11.3.10
external tables,
17.2.2
extraction, transformation, and loading (ETL),
14.1
overview,
14.1
process,
4.2
extractions
data files,
15.3.1
distributed operations,
15.3.2
full,
15.2.1
incremental,
15.2.1.2
OCI,
15.3.1.2
online,
15.2.2.1
overview,
15.1
physical,
15.2.2
Pro*C,
15.3.1.2
SQL*Plus,
15.3.1.1
F
facts,
9.1
FAST clause,
5.3.7
fast refresh,
7.1.2
restrictions,
5.3.7.1
with UNION ALL,
7.2.4
files
ultralarge,
3.2.1.1
filling gaps
with data,
18.5.6
FIRST_VALUE function,
18.2.5.1
FIRST/LAST functions,
18.3.2
FOR loops,
21.3.1
FORCE clause,
5.3.7
foreign key
constraints,
4.2.2.2
joins
snowflake schemas,
2.4.3
frequent itemsets,
18.7.4
functions
analytic,
21.3.5
COUNT,
4.1.2.1.2
CUME_DIST,
18.2.1.3
DENSE_RANK,
18.2.1.1
FIRST_VALUE,
18.2.5.1
FIRST/LAST,
18.3.2
GROUP_ID,
19.4.4
GROUPING,
19.4
GROUPING_ID,
19.4.3
LAG/LEAD,
18.2.4
LAST_VALUE,
18.2.5.1
linear regression,
18.3.5
LISTAGG,
18.3.1
NTH_VALUE,
18.2.5.2
NTILE,
18.2.1.5
PERCENT_RANK,
18.2.1.4
RANK,
18.2.1.1
ranking,
18.2.1
RATIO_TO_REPORT,
18.2.3.1
REGR_AVGX,
18.3.5.2
REGR_AVGY,
18.3.5.2
REGR_COUNT,
18.3.5.1
REGR_INTERCEPT,
18.3.5.3
REGR_SLOPE,
18.3.5.3
REGR_SXX,
18.3.5.5
REGR_SXY,
18.3.5.5
REGR_SYY,
18.3.5.5
reporting,
18.2.3
ROW_NUMBER,
18.2.1.6
WIDTH_BUCKET,
18.7.1
,
18.7.1.1
window,
21.3.5
windowing,
18.2.2
G
GROUP_ID function,
19.4.4
grouping
compatibility check,
11.1.3.3
conditions,
11.7.6
GROUPING function,
19.4
when to use,
19.4.2
GROUPING_ID function,
19.4.3
GROUPING_SETS expression,
19.5
H
hierarchical cubes,
6.2.1
,
19.10.1
in SQL,
19.10
hierarchies,
9.1
how used,
3.2.6.1
multiple,
9.2.2
overview,
3.2.6.1
rolling up and drilling down,
9.1
hints
EXPAND_GSET_TO_UNION,
11.3.8.2
,
11.7.10.4
NOWRITE,
11.7.10.1
query rewrite,
10.2
,
11.7.10.1
REWRITE,
11.7.10.1
REWRITE_OR_ERROR,
11.7.10.2
histograms
creating with user-defined buckets,
18.7.3.1
hypothetical rank,
18.3.4
I
IM column store
See
In-Memory Column Store,
2.5
indexes
bitmap indexes,
4.1.2.1.3
bitmap join,
4.1.2.2
B-tree,
4.1.3
cardinality,
4.1.2.1
nulls and,
4.1.2.1.2
partitioned tables,
4.1.2.1.3
initialization parameters
JOB_QUEUE_PROCESSES,
7.1.10
OPTIMIZER_MODE,
7.1.12
,
10.2.1
PARALLEL_MAX_SERVERS,
7.1.12
PGA_AGGREGATE_TARGET,
7.1.12
QUERY_REWRITE_ENABLED,
10.2
,
10.2.1
QUERY_REWRITE_INTEGRITY,
10.2.1
STAR_TRANSFORMATION_ENABLED,
4.5.1.1
In-Memory aggregation,
19.8.6
about,
2.7
In-Memory Column Store
about,
2.5
advantages,
2.5.1
aggregation,
19.8.6
configuring,
2.5
using,
2.5.2
VECTOR GROUP BY,
2.7.1
vector joins,
2.5.3
integrity constraints,
4.2
invalidating
materialized views,
6.4
itemsets
frequent,
18.7.4
J
JOB_QUEUE_PROCESSES initialization parameter,
7.1.10
join compatibility,
11.1.3.1
K
key lookups,
17.5.1
keys,
5.1.6
L
LAG/LEAD functions,
18.2.4
LAST_VALUE function,
18.2.5.1
level relationships,
3.2.6.1
purpose,
3.2.6.1.2
levels,
3.2.6
,
3.2.6.1.1
levels in a dimension
skipping,
9.2
limiting rows returned,
18.8
linear regression functions,
18.3.5
LISTAGG function,
18.3.1
local indexes,
4.1.2
,
4.1.2.1.3
logging
error,
17.4
logical design,
3.1
logs
materialized views,
5.4
lookup tables
See dimension tables,
5.1.6
,
5.1.6
M
manual
refresh,
7.1.6
manual refresh
with DBMS_MVIEW package,
7.1.6
MATCH_RECOGNIZE clause,
20
materialized view logs,
5.4
materialized views
aggregates,
5.2.1
altering,
6.6
analytic processing,
6.2
build methods,
5.3.4
checking status,
7.1.14
containing only joins,
5.2.2
creating,
5.3
cubes,
6.2.1
data segment compression,
5.3.3
delta joins,
11.1.3.1.3
dropping,
5.5
,
5.7
invalidating,
6.4
logs,
15.2.3.3
multiple,
11.2.8
naming,
5.3.2
nested,
5.2.3
partition change tracking (PCT),
6.1.1
partitioned tables,
7.3
partitioning,
6.1
prebuilt,
5.3
query rewrite
hints,
10.2
,
11.7.10.1
matching join graphs,
5.3.6
parameters,
10.2.1
privileges,
10.2.4
refresh dependent,
7.1.9
refreshing,
5.3.7
,
7.1
refreshing all,
7.1.8
registration,
5.5
restrictions,
5.3.6.1
rewrites
enabling,
10.2
schema design,
5.1.7
schema design guidelines,
5.1.7
security,
6.5
,
6.5
set operators,
6.2.4
storage characteristics,
5.3.3
types of,
5.2
uses for,
5.1
with VPD,
6.5.1
measures,
5.1.6
MERGE PARTITION operation,
4.4.1
MERGE statement,
7.5.1
MODEL clause,
21.1
cell referencing,
21.2.4
data flow,
21.1.1
keywords,
21.2.3
parallel execution,
21.4.1
rules,
21.2.5
modeling
logical design,
2
physical design,
2
modifying
zone maps,
13.2.3
monitoring
refresh,
7.1.13
mortgage calculation,
21.5.6
MOVE PARTITION operation,
4.4.1
multiple hierarchies,
9.2.2
multiple materialized views,
11.2.8
MV_CAPABILITIES_TABLE table,
5.8.1.2
N
nested materialized views,
5.2.3
refreshing,
7.2.3
restrictions,
5.2.3.4
net present value
calculating,
21.5.3
NEVER clause,
5.3.7
nonvolatile data,
1.1
NOREWRITE hint,
10.2
,
11.7.10.1
NTH_VALUE function,
18.2.5.2
NTILE function,
18.2.1.5
nulls
indexes and,
4.1.2.1.2
O
ON COMMIT clause,
5.3.7
ON DEMAND clause,
5.3.7
optimizations
query rewrite
enabling,
10.2
hints,
10.2
,
11.7.10.1
matching join graphs,
5.3.6
query rewrites
privileges,
10.2.4
optimizer
with rewrite,
10.1
OPTIMIZER_MODE initialization parameter,
7.1.12
,
10.2.1
ORDER BY clause,
5.3.8
outer joins
with query rewrite,
11.7.3
out-of-place refresh,
7.1.4
P
packages
DBMS_ADVISOR,
5.1.4
DBMS_DIMENSION,
9.3.2
DBMS_ERROR,
17.4.2.2
DBMS_ERRORLOG,
17.4.2.2
,
17.5.3
DBMS_MVIEW,
5.8.1.1
,
7.1
DBMS_STATS,
11.1.1
DBMS_SYNC_REFRESH,
8
parallel DML
bitmap indexes,
4.1.2
parallel execution,
4.3
parallel query
bitmap indexes,
4.1.2
PARALLEL_MAX_SERVERS initialization parameter,
7.1.12
parallelism,
4.3
partition change tracking (PCT),
6.1.1
,
7.3.1
,
11.2.7
refresh,
7.1.3
with Pmarkers,
11.2.7.4
partitioned outer join,
18.5
partitioned tables
materialized views,
7.3
partitioning,
15.2.3.2
materialized views,
6.1
prebuilt tables,
6.1.3
partitions
bitmap indexes,
4.1.2.1.3
pattern matching,
20
keywords,
20.2.2
patterns
in SQL,
20
PERCENT_RANK function,
18.2.1.4
PGA_AGGREGATE_TARGET initialization parameter,
7.1.12
physical design,
3.1
structures,
3.2.1
pivoting,
17.5.4
,
17.5.4
,
18.4
operations,
18.4
plans
star transformations,
4.5.2.2
Pmarkers
with PCT,
11.2.7.4
populating
In-Memory Column Store, In-Memory Column Store
populating,
2.5
prebuilt materialized views,
5.3
pruning
examples,
13.4.2.1
,
13.4.2.2
using zone maps,
13.4
purging data,
7.5.3
Q
query delta joins,
11.1.3.1.2
query rewrite
advanced,
11.4
checks made by,
11.1.3
controlling,
10.2.2
correctness,
10.2.3
date folding,
11.3.5
enabling,
10.2
,
10.2
hints,
10.2
,
11.7.10.1
matching join graphs,
5.3.6
methods,
11.1
parameters,
10.2.1
privileges,
10.2.4
restrictions,
5.3.6.2
using equivalences,
11.4
using GROUP BY extensions,
11.3.8.1
using nested materialized views,
11.3.1
using PCT,
11.2.7
VPD,
6.5.1.1
when it occurs,
10.1.1
with bind variables,
11.3.11
with DBMS_MVIEW package,
11.6.2
with expression matching,
11.3.10
with inline views,
11.3.2
with partially stale materialized views,
11.3.10.1
with selfjoins,
11.3.4
with set operator materialized views,
11.3.7
with view constraints,
11.3.6
QUERY_REWRITE_ENABLED initialization parameter,
10.2
,
10.2.1
QUERY_REWRITE_INTEGRITY initialization parameter,
10.2.1
R
range-partitioned table,
4.4.1
RANK function,
18.2.1.1
ranking functions,
18.2.1
RATIO_TO_REPORT function,
18.2.3.1
reference tables
See dimension tables,
5.1.6
refresh
monitoring,
7.1.13
options,
5.3.7
out-of-place,
7.1.4
partition change tracking (PCT),
7.1.3
scheduling,
7.1.15
synchronous,
8
with UNION ALL,
7.2.4
refreshing
materialized views,
7.1
nested materialized views,
7.2.3
partitioning,
7.4
zone maps,
13.3.3
REGR_AVGX function,
18.3.5.2
REGR_AVGY function,
18.3.5.2
REGR_COUNT function,
18.3.5.1
REGR_INTERCEPT function,
18.3.5.3
REGR_R2 function,
18.3.5.4
REGR_SLOPE function,
18.3.5.3
REGR_SXX function,
18.3.5.5
REGR_SXY function,
18.3.5.5
REGR_SYY function,
18.3.5.5
RELY constraints,
4.2.2.3
reporting functions,
18.2.3
restrictions
fast refresh,
5.3.7.1
nested materialized views,
5.2.3.4
query rewrite,
5.3.6.2
result set,
4.5.2
REWRITE hint,
10.2
,
11.7.10.1
,
11.7.10.1
REWRITE_OR_ERROR hint,
11.7.10.2
rewrites
hints,
11.7.10.1
parameters,
10.2.1
privileges,
10.2.4
query optimizations
hints,
10.2
,
11.7.10.1
matching join graphs,
5.3.6
rolling up hierarchies,
9.1
ROLLUP,
19.2
concatenated,
19.10.1
partial,
19.2.3
when to use,
19.2.1
root level,
3.2.6.1.1
row_limiting_clause clause,
18.8
ROW_NUMBER function,
18.2.1.6
rules
in MODEL clause,
21.2.5
in SQL modeling,
21.2.5
order of evaluation,
21.2.6
S
schemas
3NF,
2.3
design guidelines for materialized views,
5.1.7
snowflake,
2.2.1
star,
2.2.1
set operators
materialized views,
6.2.4
simultaneous equations,
21.5.4
SKIP WHEN NULL clause,
9.2
skipping levels in a dimension,
9.2
snowflake schemas,
2.4.3
complex queries,
2.4.3
source systems,
15.1
sparse data
data densification,
18.5.2
SPLIT PARTITION operation,
4.4.1
SQL modeling,
21.1
cell referencing,
21.2.4
keywords,
21.2.3
order of evaluation,
21.2.6
performance,
21.4
rules,
21.2.5
rules and restrictions,
21.3.7
staging
areas,
1.4.2
databases,
5.1
files,
5.1
STALE_TOLERATED mode,
10.2.3
star queries
star transformation,
4.5.2
star schemas
dimensional model,
2.4.1
star transformations,
4.5.2
restrictions,
4.5.2.6
STAR_TRANSFORMATION_ENABLED initialization parameter,
4.5.1.1
statistics,
11.7.9
storage
optimizing,
4.4
summary management
components,
5.1.5
sync refresh,
8
synchronous refresh,
8
T
tables
attribute-clustered,
12.1
detail tables,
5.1.6
dimension tables (lookup tables),
5.1.6
external,
17.2.2
fact tables,
5.1.6
tablespaces
transportable,
15.2.2.2
,
16.2.3
,
16.2.3.2
text match,
11.2.1
with query rewrite,
11.7.4
Third Normal Form,
2.3
time series calculations,
18.6
timestamps,
15.2.3.1
Top-N queries,
18.8
transformations,
17.1
scenarios,
17.5
SQL and PL/SQL,
17.3.1
SQL*Loader,
17.2.1
transportable tablespaces,
15.2.2.2
,
16.2.3
,
16.2.3.2
transportation
definition,
16.1
distributed operations,
16.2.2
flat files,
16.2.1
triggers,
15.2.3.3
TRUSTED mode,
10.2.3
types
attribute clustering,
12.1.1
zone maps,
13.1.3
U
ultralarge files,
3.2.1.1
unique
constraints,
4.2.2.1
identifier,
2.2
,
3.2
update frequencies,
5.1.9
UPDATE statement,
21.2.8
update windows,
5.1.9
UPSERT ALL statement,
21.2.8
UPSERT statement,
21.2.8
using
In-Memory Column Store,
2.5.2
V
validating dimensions,
9.5
VECOTR GROUP BY aggregation
scenarios,
2.7.2
VECTOR GROUP BY
aggregation,
19.8.6
VECTOR GROUP BY aggregation
IM column store,
2.7.1
optimizing star queries,
4.5.4
usage in queries,
2.7.3
vector joins
In-Memory Column Store,
2.5.3
view constraints,
4.2.2.7
,
11.3.6
VPD
and materialized views,
6.5.1
restrictions with materialized views,
6.5.1.2
W
WIDTH_BUCKET function,
18.7.1
,
18.7.1.1
window functions,
21.3.5
windowing functions,
18.2.2
Z
zoen maps
about staleness,
13.3.1
zone maps,
13.1
about,
13.1
about refreshing,
13.3.2
automatic refresh,
13.1.6.2
benefits,
13.1.4
compiling,
13.2.5
creating,
13.2.2.2
with attribute clustering,
13.2.2.1
data dictionary views,
13.5
dropping,
13.2.4
maintaining,
13.2.7
maintenance,
13.1.6.1
modifying,
13.2.3
privileges,
13.2.1
pruning,
13.4
refreshing,
13.3.3
types,
13.1.3
usage for SQL statements,
13.2.6.2
usage for SQL workloads,
13.2.6.1
usage scenarios,
13.1.5
with attribute clustering,
13.1.2
Scripting on this page enhances content navigation, but does not change the content in any way.