Ho Chi Minh City University of Technology

Faculty of Computer Science and Engineering

Online Instructions for

Chapter 2: Divide-and-conquer

Algorithms Analysis and Design

(CO3031)

Instructor: Dr. Vo Thi Ngoc Chau

Email: chauvtn@hcmut.edu.vn

Semester 2 – 2019-2020
 Course Outline
 Chapter 1. Fundamentals
 Chapter 2. Divide-and-Conquer Strategy
 Chapter 3. Decrease-and-Conquer Strategy
 Chapter 4. Transform-and-Conquer Strategy
 Chapter 5. Dynamic Programming and Greedy
Strategies
 Chapter 6. Backtracking and Branch-and-Bound
 Chapter 7. NP-completeness
 Chapter 8. Approximation Algorithms
2
 Ho Chi Minh City University of Technology
Faculty of Computer Science and Engineering

Online Instructions for

Chapter 2: Divide-and-conquer

PART 2

Semester 2 – 2019-2020
 Chapter 2.
Divide-and-Conquer Strategy

 2.1. Divide-and-conquer strategy

 2.2. Quicksort

 2.3. Mergesort

 2.4. External sorting

 2.5. Binary search tree

4
 2.3. Mergesort
 To sort a given array of n elements, divide it in half,
sort the two halves (recursively), and then merge the
two halves together.
 Merging: given two sorted arrays a[1..M] and b[1..N],
we wish to merge them into the third sorted array
c[1..M+N].
i:= 1; j :=1;
for k:= 1 to M+N do
if a[i] < b[j] then
begin c [k] := a[i]; i:= i+1 end
else begin c[k] := b[j]; j := j+1 end;
5
 2.3. Mergesort
procedure mergesort(left,right: integer);
var i, j, k, m : integer;
begin
if right-left>0 then
begin
m:=(right+left)/2; mergesort(left,m); mergesort(m+1,right);
for i := m downto left do b[i] := a[i];
for j :=m+1 to right do b[right+m+1-j] := a[j];
//i := left; j := right;
for k :=left to right do Merging on b:
if b[i] < b[j] then
0 6 7 11 9 4
begin a[k] := b[i] ; i := i+1 end i j
else begin a[k] := b[j]; j:= j-1 end;
end;
end; 6
2.3. Mergesort

7
 2.3. Mergesort

 Time complexity for sorting n elements

 Abstract operation: comparison

 All the cases:

C1 = 0
Cn = 2Cn/2 + n when n≥2
Cn = nlog2n = O(nlog2n)
NOTE: Mergesort needs memory space of n elements for merging!
8
 2.4. External sorting
 Sorting the large files stored in secondary storage is
called external sorting. External sorting is very
important in database management systems.

 When estimating the computational time of the

algorithms that work on files in hard disks, we must
consider the number of times we read a block to main
memory or write a block to secondary storage.

 Such operation is called block access or disk access.

9
 2.4. External sorting

 The most commonly used technique for external

sorting is the external sort-merge algorithm.
 This external sorting method has two stages.
 create runs: br = number of blocks, M = buffer size
 br/M runs from any in-place in-memory sorting algorithm

 merge runs: one block for output, each block of the

rest for each input run
 log(M-1)br/M merging phases
10
 2.4. External sorting
 create runs: br/M runs from in-place in-memory sorting
i = 0;
repeat
read M blocks of the file, or the rest of the file, whichever is
smaller;
sort the in-memory part of the file;
write the sorted data to the run file Ri;
i := i+1;
until the end of the file;

 merge runs: one block for output, each block of the rest
for each input run with log(M-1)br/M merging phases
11
 2.4. External sorting
 create runs: br/M runs from in-place in-memory sorting
 merge runs: one block for output, each block of the rest
for each input run with log(M-1)br/M merging phases
read one block of each of the N files Ri into the buffer; //N = br/M
repeat
choose the first tuple (in sort order) among all buffer pages;
write the tuple to the output and delete it from the buffer page;
if the buffer page of any run Ri is empty and not end-of-file(Ri)
then read the next block of Ri into the buffer page;
until all buffer pages are empty;
12
 2.4. External sorting
 File: 2, 3, 9, 1, 5, 6, 0, 9, 10, 16, 8, 11, 2,
15, 9, 4, 2, 0, 7, 4, 5, 3, 12, 13, 1, 0
 1 record/block, 26 records => br=26 blocks
 M = 4 blocks
 Number of runs = br/M = 7
 Number of merging phases=logM-1br/M=2

2,3,9,1, 5,6,0,9, 10,16,8,11, 2,15,9,4, 2,0,7,4, 5,3,12,13, 1,0

Generate runs

1,2,3,9, 0,5,6,9, 8,10,11,16, 2,4,9,15, 0,2,4,7, 3,5,12,13, 0,1

13
 2.4. External sorting
 File: 2, 3, 9, 1, 5, 6, 0, 9, 10, 16, 8, 11, 2, 15, 9, 4, 2, 0, 7, 4,
5, 3, 12, 13, 1, 0

 1 record/block, 26 records => br=26 blocks

 M = 4 blocks
 Number of runs = br/M = 7
 Number of merging phases = logM-1br/M = 2

1,2,3,9, 0,5,6,9, 8,10,11,16, 2,4,9,15, 0,2,4,7, 3,5,12,13, 0,1

Merge in phase 1

0,1,2,3,5,6,8,9,9,10,11,16, 0,2,2,3,4,4,5,7,9,12,13,15, 0,1

Merge in phase 2

0,0,0,1,1,2,2,2,3,3,4,4,5,5,6,7,8,9,9,9,10,11,12,13,15,16
14
 2.4. External sorting

 Time complexity

 Abstract operation: block access

 All the cases:

2br + 2br log M-1 br/M  = 2br(log M-1 br/M  +1)

br = number of blocks,
run generation run merging M = buffer size

15
 Algorithms Analysis and Design
(CO3031)

 Chapter 2. Divide-and-conquer
 2.1. Divide-and-conquer strategy

 2.2. Quicksort

 2.3. Mergesort

 2.4. External sorting

 2.5. Binary search tree

16