Просмтор этого файла доступен только зарегистрированным пользователям. Но у нас супер быстрая регистрация: достаточно только электронной почты!

Текст из файла (страница 3)

PARAMETER (L=1200, ITMAX=4)

REAL A(L,L), EPS, MAXEPS, B(L,L)

CHPF$ PROCESSORS P(2,2)

CHPF$ DISTRIBUTE A ( BLOCK, BLOCK) ONTO P

CHPF$ ALIGN B( I, J ) WITH A( I, J )

C arrays A and B with block distribution

PRINT *, '********** TEST_JACOBI **********'

MAXEPS = 0.5E - 7

CDVM$ PARALLEL (J,I) ON A(I, J)

C nest of two parallel loops, iteration (i,j) will be executed on

C processor, which is owner of element A(i,j)

DO 1 J = 1, L

DO 1 I = 1, L

A(I, J) = 0.

IF(I.EQ.1 .OR. J.EQ.1 .OR. I.EQ.L .OR. J.EQ.L) THEN

B(I, J) = 0.

ELSE

B(I, J) = ( 1. + I + J )

ENDIF

1 CONTINUE

DO 2 IT = 1, ITMAX

EPS = 0.

CDVM$ PARALLEL (J, I) ON A(I, J), REDUCTION ( MAX( EPS ))

C variable EPS is used for calculation of maximum value

DO 21 J = 2, L-1

DO 21 I = 2, L-1

EPS = MAX ( EPS, ABS( B( I, J) - A( I, J)))

A(I, J) = B(I, J)

21 CONTINUE

CDVM$ PARALLEL (J, I) ON B(I, J), SHADOW_RENEW (A)

C Copying shadow elements of array A from

C neighboring processors before loop execution

DO 22 J = 2, L-1

DO 22 I = 2, L-1

B(I, J) = (A( I-1, J ) + A( I, J-1 ) + A( I+1, J)+

* A( I, J+1 )) / 4

22 CONTINUE

PRINT *, 'IT = ', IT, ' EPS = ', EPS

IF ( EPS . LT . MAXEPS ) GO TO 3

2 CONTINUE

3 CONTINUE

C OPEN (3, FILE='JACOBI.DAT', FORM='FORMATTED')

C WRITE (3,*) B

C CLOSE (3)

END

INTERVAL ( NLINE=8 SOURCE=jac.fdv ) LEVEL=0 EXE_COUNT=1

--- Main characteristics ---

Note: there are only non-distributed data print operators in the program, this fact explains the absence of input/output communication losses. As such data have the same value on each processor such operations are executed by input/output processor without interprocessor exchanges.

--- Comparative characteristics ---

7Appendix. The list of characteristics

7.1Main characteristics and their components

1. Efficiency coefficient (Parallelization efficiency) is ratio of productive time to total processor time.

2. Time of execution (Execution time).

3. The number of used processors (Processors).

4. Total processor time (Total time) is production of the time of execution (Execution_time) by the number of used processors (Processors).

5. Productive time (Productive time) is the sum of productive processor time (CPU), input/output time (I/O) and productive system time (Sys).

6. Lost time (Lost time).

7. Insufficient parallelism (Insufficient par) and its components.

8. Communications and all components (Communication).

9. Idle (Idle time).

10. Imbalance (Load Imbalance).

11. Potential synchronization losses and all components (Synchronization).

12. Potential time variation losses and all components (Time variation).

13. Overlapping time and all components (Overlap).

7.2Characteristics of program execution on each processor

1. Lost time (Lost time) is the sum of insufficient parallelism losses (User Insufficient par), system insufficient parallelism losses (Sys Insufficient par), communications losses (Communication) and idle (Idle).

2. Insufficient parallelism losses (User insufficient par).

3. System insufficient parallelism losses (Sys insufficient par).

4. Time of losses because of the given processor idle (Idle time) is difference between maximal interval execution time (on any processor) and interval execution time on the given processor.

5. Total communication time (Communication).

6. Real time of losses because of dissynchrinization (Real synchronization).

7. Potential time of losses because of dissynchrinization (Synchronization).

8. Potential time of losses because of time variation (Variation).

9. Time of asynchronous operation overlapping (Overlap).

10. Losses because of load imbalance (Load Imbalance) is difference between maximal processor time (CPU + Sys) and the time on the given processor.

11. Time of interval execution (Execution time).

12. Productive processor time (User CPU time).

13. Productive system time (Sys CPU time)

14. Input/output time (I/O time).

15. Time of collective operation start (Start operation).

16. Number of processors used for interval (Processors).

17. Communication times for all types of collective operations (Reduction, Shadow, Remote access, Redistribution и I/O), besides times of their start.

18. Real dissynchronization losses for all types of collective operations.

19. Potential dissynchronization losses for all types of collective operations.

20. Potential time variation losses for all types of collective operations.

21. Time of overlapping for all collective operations (Overlap).

Характеристики

Список файлов учебной работы