CMU Intro to Database Systems

Question 1

why do we not use flat file system for storage?

Answer 1

we cannot ensure the integrity of the database and by that we can find ourselves changing our application code multiple times

Question 2

what are the relational model pillars?

Answer 2

1 - structure : the definition of the database relations and their contents
2 - integrity : ensures the database satisfy constraints
3 - manipulation : programming interface for accessing and modifying a database contents

Question 3

does database store it’s data on one file or more?

Answer 3

more than one file to not reach the max limit of file size on OS

Question 4

what is a storage manager?

Answer 4

it is responsible for maintaining database files

Question 5

what does a storage manager do?

Answer 5

it organizes database files as a collection pages

• tracking data read/written pages
• tracking the available space

Question 6

what is a page?

Answer 6

it is a fixed sized of a block of data

Question 7

what are the characteristics of a page?

Answer 7

1 - it may contain anything from indexed, metadata , tuples
2 - most systems do not mix page types
3 - some systems required that each page is self contained meaning each page will have the meta data sufficient to interpret the page itself

Question 8

what does the DBMS do to map page ids to physical locations?

Answer 8

an indirection layer

Question 9

what are the notions of pages?

Answer 9

1 - hardware page
2- OS page
3 - Database Page

Question 10

what are the ways that a DBMS organize pages on a disk?

Answer 10

1 - Heap Organization
2 - sequential or sorted file organization
3 - hashing file organization

Question 11

what is heap file organization?

Answer 11

is an unordered collection of pages where tuples are stored in random order

Question 12

why is heap file organization unordered?

Answer 12

because we use SQL to get the data so the data do not need to be ordered for us to get the data

Question 13

what are the ways to represent heap file organization?

Answer 13

1 - linked list

2 - page directory

Question 14

how do we use linked list to represent heap file organization?

Answer 14

• we have a header that has two pointers
  1 - a pointer to free header list
  2 - a pointer to a data header list
  3 - each list is double linked so we can go forward and back
  4 - each page contains the amount of free slots it has

Question 15

do we use linked list or page directory for heap file organization?

Answer 15

we use page directory

Question 16

how do we use a page directory to represent heap file?

Answer 16

the directory keep track of the empty pages and data pages as a map

Question 17

what does a page have?

Answer 17

1 - header

2 - data

Question 18

what does a header of a page have?

Answer 18

meta data about the page
1 - checksum
2 - DBMS version
3- page size

Question 19

how are the data stored in a page?

Answer 19

1 - tuple oriented

2 - log structured