Scribd
Upload
7 views8 pages

Parallel Processing Performance Metrics

The document contains graphs that show how execution time, speedup, and efficiency change as the number of processors is increased for two parallel programs using OpenMP (OMP) and MPI. Execution time decreases as more processors are used for both programs, though MPI sees greater speedup. Speedup increases with more processors but levels off around 8-16 processors. Efficiency decreases as more processors are used due to overhead costs, dropping below 0.5 for 64 processors with MPI.

Uploaded by

Nur Fatini
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
7 views8 pages

Parallel Processing Performance Metrics

The document contains graphs that show how execution time, speedup, and efficiency change as the number of processors is increased for two parallel programs using OpenMP (OMP) and MPI. Execution time decreases as more processors are used for both programs, though MPI sees greater speedup. Speedup increases with more processors but levels off around 8-16 processors. Efficiency decreases as more processors are used due to overhead costs, dropping below 0.5 for 64 processors with MPI.

Uploaded by

Nur Fatini
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd

Num of processor Execution time OMP Execution time OMP

1 0.042 0.045
2 0.028 0.042
0.04
4 0.026
0.035
8 0.024
16 0.023 0.03
0.028
32 0.014 0.025 0.026
0.024 0.023
64 0.005 0.02
0.015 0.014
0.01
0.005 0.005
0
1 2 4 8 16 32 64

Num of processor Speedup


1 1 Speedup
2 1.536 9
4 1.608 8.368
8
8 1.73
7
16 1.816
6
32 3.055
64 8.368 5
4
3 3.055

2 1.816
1.536 1.608 1.73
1 1
0
1 2 4 8 16 32 64

Number of processor Efficiency


1 1
Efficiency
2 0.768 1.2
4 0.402 1
1
8 0.216
16 0.114 0.768
0.8
32 0.095
64 0.131 0.6

0.402
0.4
0.216
0.2 0.114 0.131
0.095

0
1 2 4 8 16 32 64
0.6

0.402
0.4
0.216
0.2 0.114 0.131
0.095

0
1 2 4 8 16 32 64
P

0.023

0.014

0.005

16 32 64

8.368

3.055

2 64

0.131
95

2 64
0.131
95

2 64
Num of processor Execution time MPI
1 3.752 Execution time MPI
2 1.879 4 3.752
4 0.938
3.5
8 0.472
16 0.241 3

32 0.145 2.5
64 0.112 1.879
2

1.5
0.938
1
0.472
0.5 0.241 0.145 0.112
0
1 2 4 8 16 32 64

Num of processor Speed Up MPI


1 1 Speed Up MPI
2 1.997 40
4 3.998 35 33.501
8 7.951
30
16 15.59 25.911
32 25.911 25

64 33.501 20
15.59
15

10 7.951
3.998
5 1.997
1
0
1 2 4 8 16 32 64

Num of processor Efficiency MPI


1 1 Efficiency MPI
2 0.998 1.2
4 0.99954 1 0.998 0.99954 0.994 0.974
8 0.994 1

16 0.974 0.81
0.8
32 0.81
0.6 0.523

0.4
1 0.998 0.99954 0.994 0.974
1
0.81
0.8

64 0.523 0.6 0.523

0.4

0.2

0
1 2 4 8 16 32 64
PI

241 0.145 0.112

6 32 64

33.501

911

2 64

81

0.523
81

0.523

2 64

You might also like