データベースの処理に関する次の記述中のa,bに入れる字句の適切な組合せはどれか。 データベースに対する処理の一貫性を保証するために,関連する一連の処理を一つの単位にまとめて処理することを[ a ]といい,[ a ]が正常に終了しなかった場合に備えて[ b ]にデータの更新履歴を取っている。
平成29年度 秋期 ITパスポート試験 問76
This question's explanation and your answers so far.
問 76
Incorrectア
a’s “Normalization” is wrong. Normalization is a technique in table design for eliminating data duplication and inconsistency, not “grouping a series of operations into a single unit.” Also, b’s “Backup file” is not the storage location for the update history, so both a and b are inappropriate.
Incorrectイ
b’s “Log file” is correct, but a’s “Normalization” is wrong. Normalization is a design technique for organizing data structure, not the concept of grouping operations into a single unit. Since a does not match, it is inappropriate.
Incorrectウ
a’s “Transaction processing” is correct, but b’s “Backup file” is wrong. A backup file is a copy of the data at a certain point in time (a static copy) and does not record the sequential update history. The storage location for keeping the update history is the log file, so b does not match and is inappropriate.
- エ
正解 The correct answer is エ (a: Transaction processing / b: Log file). Statement a exactly matches the description “grouping a series of related operations into a single unit and processing them together in order to guarantee the consistency of processing on a database,” and this is called transaction processing. A transaction is treated as a unit that either succeeds entirely or is not reflected at all (atomicity). b is the storage location that keeps the “data update history” in preparation for the case where the transaction does not complete normally, and this is the log file (also called a journal file). By recording the values before and after each update in time order, recovery is possible when a failure occurs, through rollback (returning to the state before the update) or roll-forward (advancing to the state after the update). エ, in which both a and b match, is correct.
- A transaction is a mechanism that groups multiple related operations and treats them as a single indivisible unit in order to maintain database consistency (atomicity: either everything is reflected or nothing is reflected at all).
- A log file (journal) records the history before and after updates and is used for recovery through rollback / roll-forward when a failure occurs. Its role differs from that of a backup file, which is a copy at a certain point in time.