| 
|
Topic: database consistency and reliability
topics >
computer science >
consistency testing
Topic:
constraints
Topic:
database change management
Topic:
database implementation
Topic:
database security
Topic:
database transactions
Topic:
dependency analysis
Topic:
error safe systems
Topic:
file system reliability
Topic:
log-structured rollback-recovery
Topic:
logging data and events
Topic:
managing changes in hypertext
Topic:
reliability of distributed systems
Topic:
replicated data
Topic:
software change management
Topic:
undefined, null, and other signal values
Topic:
updating information in a distributed system
Summary
A database is consistent if it satisfies stated and unstated assertions about its data. It is reliable if the data is accessible, current, and consistent. A database schema may define constraints that new data must satisfy. For example, a foreign-key column matches a row in the corresponding table.Imported data often contains incorrect and missing data. The data may require manual cleansing or normalization. A null value is commonly used for missing data. (cbb 8/06)
Subtopic: consistency audit
Quote: a consistent database is one that satisfies all stated assertions about its data [»shavMJ2_1980] Quote: conventional databases are a snapshot of the current state; may be somewhat out-of-date [»snodR9_1986] Quote: a database schema needs a consistent definition of all terms; does not exist, so use exceptions [»borgA10_1986] Quote: a database model should allow arbitrary combinations of constraints on relationships [»kentW_1978]
Subtopic: data cleansing
Quote: audit and correct the relational databases for 5ESS telephone switches; e.g., if a unit is out-of-service, its children are either out-of-service or in a transient state [»guptNK1_1997]
Subtopic: insuring consistency
Quote: interactive data cleaning system that integrates transformation and discrepancy detection [»ramaV9_2001] Quote: transformations shown at once on visible records; like a spreadsheet Quote: optimize recall, precision, and length when parsing a data domain [»ramaV9_2001] Quote: use minimum description length to select how to parse data in a column
Subtopic: metadata
Quote: an information system depends on accurate, consistent data; constraints defined as part of the databases schema [»borgA10_1986] Quote: electronic data publishing can revise or correct data after publication; difficult in print media [»cinkMJ5_1991] Quote: GenBank checks sequence data far more extensively then reviewers do; many automatic checks [»cinkMJ5_1991]
Subtopic: avoid duplicate data
Quote: implement atomic update of metadata information by updating a duplicate page and swapping page pointers when done [»chenPM9_1996]
Subtopic: distributed update
Quote: keeping two related, but different sets of data is asking for trouble; sooner or later someone will get them out of sync [»koenA12_1995]
Quote: Bayou supports four session guarantees: read your writes, monotonic reads, writes follow reads, and monotonic writes; allows self consistency
[»edwaWK10_1997]
Topic: consistency testing (60 items)
Topic: constraints (35 items)
Topic: database change management (12 items)
Topic: database implementation (18 items)
Topic: database security (12 items)
Topic: database transactions (27 items)
Topic: dependency analysis (34 items)
Topic: error safe systems (76 items)
Topic: file system reliability (26 items)
Topic: log-structured rollback-recovery (13 items)
Topic: logging data and events (17 items)
Topic: managing changes in hypertext (21 items)
Topic: reliability of distributed systems (35 items)
Topic: replicated data (51 items)
Topic: software change management (48 items)
Topic: undefined, null, and other signal values (34 items)
Topic: updating information in a distributed system
(50 items)