1. Thesis: On the development of numerical parallel algorithms for the insetting procedure Master of Science in Communication & Information Systems Department of Informatics & Communications TEI of Central Macedonia Christos Christodoulou Supervisor D.Varsamis Lecturer

2. Objectives  Find the optimal number of insets  Find insets with numerical Logic  The algorithm can be used not only at maps

3. Cartography  From Greek χάρτης hartes, "map" and γράφειν graphein, "write") is the study and practice of making maps.  Map construction is one of the oldest human activities.  According to archaeologists older projects have been and could still qualify maps dating to 30,000 years ago

4. Island Cartography  Deals with special cartographic problems  The need of inset map creation for very small islands  sometimes isolated ones  Must be displayed in the main map  Key factor for insetting in Island is the “complexity of land discontinuity”

5. Example Map Chalkidiki

6. Existing Method Algorithm  Calculate the Q correlation  Uses standard frames  Search step by step all the map Example

7. Step 1

8. Step 2

9. Step 3

10. Numerical Algorithm  Find insets without searching with standard frames  Calculate the maximum Q correlation that the selection area can be maximized  Uses only addition of the map pixels

11. Numerical Algorithm example 1 Example map Rasterized Map

12. Numerical Algorithm example 2

13. Numerical Algorithm example 3

14. Numerical Algorithm example 4

15. Parallel implementation

17. Implementation Cases  Many ways and execution scenarios  SPMD  Tasks  Examine two scenarios  N ( accuracy ) according to number of workers  N (accuracy ) has the maximum value

18. Case 1 N = Workers mapNLabsTime secInsetsqInset Dim. 50.tif 884.442920250x50 50.tif 16 5.506292.29750x50 50.tif 32 7.418292.28750x50 50.tif 64 11.941292.28350x50 As we can see at the results the time increases as we increase the number of workers instead of reducing. This is happened because we have delay for the communication time.

19. Workers - Time

20. Case 2 N stable mapNLabsTime secInsetsqInset Dim. 50.tif 64120.761292.28350x50 50.tif 64418.830292.28350x50 50.tif 64811.369292.28350x50 50.tif 64168.485292.28350x50 50.tif 64328.399292.28350x50 50.tif 64 11.359292.28350x50

21. Workers-Time ( N stable)

22. N stable Speed Up - Efficiency 1-core4-cores8-cores16-cores32-cores64-cores S 1.00001.10251.82612.44672.47181.8277 E 1.00000.27560.22820.15290.07720.1848

23. Workers-Speed Up ( N stable)

24. Workers-Efficiency N stable

25. Conclusion  The theoretical computational time of the numerical parallel algorithm is almost the same with the logical parallel algorithm.  The efficiency of the numerical algorithm is in general very good, helping the cartographers who use the algorithm to reduce its execution time in a local machine with multicore processor or to a grid computer.

26. Future Work  Use different methods for parallel implementation  SMPMD  Tasks  Transform the algorithm to a web-based application in parallel architecture  distributed memory  in a network of computers or in a grid computer