《数据库理论与技术》期终考试复习题

一、填空题：

1. 数据仓库是面向(主题的)、(集成的) 、(相对稳定的)、(反映历史变化)的数据集合.
2. 多维数据模型由(维表)和(事实表)定义，其常见的形式有(星型)、(雪花型) 及(事实星座型)；在多维立方体的某一维上选定一个维成员值的操作称为（切片）。
3. 在ORDB中，从已知的属性值找未知的属性值时沿途经过的属性名构成的式子称为(路径表达式)；将一个嵌套关系转换为1NF的过程称为(解除嵌套)；在ODMG对象模型中，由类中所有持久对象构成的集合称为(类外延)；
4. 描述时间数据的最小时间单位称为(时间粒度)；而(系统时间量子(Chronon))是计算机系统所支持的最小的、不可分割的时间间隔。时态数据模型中的三种时间分别是（用户自定义时间(user-defined time)）、（有效时间(valid time)）、（事务时间(transaction time)）。
5. 分布式数据库的六级模式结构分别是（全局外模式）、（全局概念模式）、（分片模式）、（分配模式）、（局部概念模式）、（局部内模式）；分布式数据库系统中，查询代价主要由（CPU代价）、（I／O代价）、（通信代价）三部分构成。
6. 两关系间的连接运算可通过（嵌套循环联接）、（块嵌套循环联接）、索引嵌套循环联接
7. (归并联接)及（散列联接）、等不同方法实现，当采用块嵌套循环方法时，通常将较大的关系作为嵌套循环计算过程中的(内)关系。
8. 线性散列是为了解决可扩展散列面临的桶的（桶的数目过快增长，但其利用率不足）及（ ）问题而引入的。
9. 乐观的并发控制协议中，事务的执行可划分为（读与执行阶段）、（有效性检查阶段）、（写阶段）等三个阶段。
10. 影响并行数据库系统性能的3个因素分别是（启动代价）、（干扰）和（倾斜）。
11. 在并行数据库中，将互相独立的多个操作或者一个操作内互相独立的多个子操作分配给不同的处理机并行执行称为（流水线（水平）并行），将存在流水线方式依赖关系的多个操作分配给不同处理机并行执行称为（独立（垂直）并行）。
12. CAP定理中的C、A、P分别是指（一致性）、（可用性）、（分区容忍性）。
13. 区块链是一个（去中心化）、（去信任）的公开透明分布式账本。
14. 按照开放程度，区块链可分为（公有链）、（私有链）和（联盟链）等三种类型；区块链的基础架构模型包括（应用层）、（合约层）、（激励层）、（共识层）、（网络层）、（数据层）等六层。

二、选择题:

1. The separation of the data definition from the program is known as: ______________b
   a) Data dictionary
   b) Data independence
   c) Data integrity
   d) Referential integrity

2. In order to reduce the overhead in retrieving the records from the storage space we use: ____________b
   a) Logs
   b) DB buffer
   c) Medieval space
   d) Lower records
   Answer: b
   Explanation: The output to stable storage is in units of blocks.

3. The main reason why data is retrieved in “blocks” is to: a
   a) minimize the number of actual disk accesses
   b) save space
   c) make programming easier
   d) increase the lifetime of hardware

4. The order of log records in the stable storage ____________ as the order in which they were written to the log buffer.a
   a) Must be exactly the same b) Can be different c) Is opposite
   d) Can be partially same
   Answer: a
   Explanation: As a result of log buffering, a log record may reside in only main memory (volatile storage) for a considerable time before it is output to stable storage.

5. The difference between files storing spanned(跨块的)versus unspanned (非跨块的)records is that：______________
   a) a file with spanned records will use less disk space for storing records than with unspanned records, if an integral number of records do not fit in a block.
   b) a file with spanned records can have records that are stored on more than one disk block.
   c) a file with spanned records must be used when the size of a record is larger than the block size.
   d) all of the above. e) NOA.

6. An index is used in relational database systems to：______________b
   a) improve the efficiency of normalizing relations.
   b) improve the efficiency of retrieving records from a relation.
   c) improve the efficiency of the Create Table statement.
   d) b and c only. e) NOA.

7. Which of the following would not be considered a benefit of indexing? ____________b
   a) To improve performance during data sorting.
   b) To insure uniqueness of key values.
   c) To improve performance during large sequential table scans.
   d) To avoid reading the records when processing queries that retrieve only indexed columns.
   e) To help query optimizer in cost estimation.

8. The difference between a dense(稠密) index and a sparse(稀疏) index is that: __________c
   a) a dense index contains keys and pointers for a subset of the records whereas a sparse index contains keys and pointers for every record.
   b) a dense index can only be a primary index whereas a sparse index can only be a secondary index.
   c) a dense index contains keys and pointers for each record whereas a sparse index contains keys and pointers for a subset of the records.
   d) the size of dense index is always smaller than the size of sparse index.

9. In ordered indices the file containing the records is sequentially ordered, a ___________ is an index whose search key also defines the sequential order of the file.
   a) Clustered index b) Structured index c) Unstructured index d) Nonclustered