Öz
With the developing multi-core technology, utilization of parallel computing approach for the high performance algorithms become widespread. In this study, analysis of parallel image filter that was realized with Java threads on multicore computer has been presented. Experimental results were obtained using dual core and quad core processor with Hyper Threading technology. For this purpose comparative speed-up and parallel efficiency graphs have been plotted referencing the single core single core computation results. According to experiments, dual core processor using two threads produced about 1.9 fold speed-up while four threads produced about 2.9 fold speed-up and quad core processor using four threads produced about 3.6 fold speed-up while eight threads produced about 5.7 fold speed-up has been obtained.
Anahtar Kelimeler
Image Processing Parallel Computing Java Threads Multi-core Processor.
Kaynakça
- R. C. Gonzalez, R. E. Woods,( 2008). Digital Image Processing,Prentice Hall.
- Babic, Z.V., (2003). An efficient noise removal and edge preserving convolution filter 6th International Conference on Telecommunications in Modern Satellite, Cable and Broadcasting Service, 2, 538 - 541 vol.2
- S. Yu, M. Clement, Q. Snell, B. Morse, (1998) “Parallel algorithms for image convolution”, Proceedings of the International Conference on Parallel and Distributed Techniques and Applications, Las Vegas, Nevada.
- J. Kepner, “A Multi-Threaded Fast Convolver for Dynamically Parallel Image filtering”, Journal of Parallel and Distributed Computing, Vol. 63, pp. 360–372, 2003
- S. Nakariyakul, “Fast spatial averaging: an efficient algorithm for 2D mean filtering”, The Journal of Supercomputing, DOI: 10.1007/s11227-011-0638-9, Online, 2011
- P. Frost Gorder, (2007). Multicore Processors for Science and Engineering, IEEE Computing in Science & Engineering, 9, 3-7.
- G. Andrews, (2000). Foundations of Multithreaded, Parallel, and Distributed Programming, Addison-Wesley
- B. Sanden, (2004). Coping with Java threads, IEEE Computer, 37, 20-27
- D. Lea, (1997). Concurrent Programming in Java: Design Principles and Patterns, AddisonWesley.
- Stallings (2005). Operating Systems, Internals and Design Principles. Pearson: Prentice Hall
- K. K. Yue, D. J. Lilja, “Parallel Loop Scheduling for High-Performance Computers”, High-Performance Parallel Computing Research Group Technical Report No. HPPC-94-13, 1994
- Z. Fang, P. Tang, P.C. Yew, C. Q. Zhu, “Dynamic processor self-scheduling for general parallel nested loops”, Vol. 39, pp. 919 – 929, 1990
- Z. Juhasz, (1998). “An Analytical Method for Predicting the Performance of Parallel Image Processing Operations”, The Journal of Supercomputing, Vol.12, pp.157-174
Öz
Çok çekirdekli işlemci teknolojisinin gelişmesiyle birlikte yüksek işlem gücü gerektiren birçok algoritmanın hızlandırılması için paralel hesaplama yaklaşımının kullanımı yaygın hale gelmiştir. Sunulan çalışmada, Java iş parçacıkları ile gerçekleştirilen paralel görüntü filtresinin çok çekirdekli işlemci üzerinde başarım analizi gerçekleştirilmiştir. Deneysel sonuçlar, iki çekirdekli ve dört çekirdekli, Hyper Threading teknolojisini destekleyen işlemcilere sahip bilgisayarlar kullanılarak elde edilmiştir. Bu amaçla tek çekirdekli hesaplama sonuçları referans alınarak hızlandırma ve paralel verimlilik grafikleri karşılaştırmalı olarak elde edilmiştir. Sonuçlar yaklaşık olarak iki çekirdekli işlemcide iki iş parçacığı ile 1,9 kat dört iş parçacığı ile 2,9 kat, dört çekirdekli işlemcide dört iş parçacığı ile 3,6 kat, sekiz iş parçacığı ile 5,7 kata kadar hızlandırma elde edilmiştir.
Anahtar Kelimeler
Görüntü İşleme Paralel Hesaplama Java İş Parçacıkları Çok Çekirdekli İşlemci.
Kaynakça
- R. C. Gonzalez, R. E. Woods,( 2008). Digital Image Processing,Prentice Hall.
- Babic, Z.V., (2003). An efficient noise removal and edge preserving convolution filter 6th International Conference on Telecommunications in Modern Satellite, Cable and Broadcasting Service, 2, 538 - 541 vol.2
- S. Yu, M. Clement, Q. Snell, B. Morse, (1998) “Parallel algorithms for image convolution”, Proceedings of the International Conference on Parallel and Distributed Techniques and Applications, Las Vegas, Nevada.
- J. Kepner, “A Multi-Threaded Fast Convolver for Dynamically Parallel Image filtering”, Journal of Parallel and Distributed Computing, Vol. 63, pp. 360–372, 2003
- S. Nakariyakul, “Fast spatial averaging: an efficient algorithm for 2D mean filtering”, The Journal of Supercomputing, DOI: 10.1007/s11227-011-0638-9, Online, 2011
- P. Frost Gorder, (2007). Multicore Processors for Science and Engineering, IEEE Computing in Science & Engineering, 9, 3-7.
- G. Andrews, (2000). Foundations of Multithreaded, Parallel, and Distributed Programming, Addison-Wesley
- B. Sanden, (2004). Coping with Java threads, IEEE Computer, 37, 20-27
- D. Lea, (1997). Concurrent Programming in Java: Design Principles and Patterns, AddisonWesley.
- Stallings (2005). Operating Systems, Internals and Design Principles. Pearson: Prentice Hall
- K. K. Yue, D. J. Lilja, “Parallel Loop Scheduling for High-Performance Computers”, High-Performance Parallel Computing Research Group Technical Report No. HPPC-94-13, 1994
- Z. Fang, P. Tang, P.C. Yew, C. Q. Zhu, “Dynamic processor self-scheduling for general parallel nested loops”, Vol. 39, pp. 919 – 929, 1990
- Z. Juhasz, (1998). “An Analytical Method for Predicting the Performance of Parallel Image Processing Operations”, The Journal of Supercomputing, Vol.12, pp.157-174
Ayrıntılar
| Birincil Dil | Türkçe |
|---|---|
| Bölüm | Makaleler |
| Yazarlar | |
| Yayımlanma Tarihi | 1 Mart 2013 |
| Yayımlandığı Sayı | Yıl 2013 Cilt: 2 Sayı: 1 |