Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Artículo científico |
| Lingua: | en |
| Pubblicazione: |
Universidad Nacional de Colombia
2009
|
| Soggetti: | |
| Accesso online: | https://www.redalyc.org/articulo.oa?id=133112608024 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866569317098717184 |
|---|---|
| author | Mauricio Alvarez |
| author_facet | Mauricio Alvarez |
| contents | Performance evaluation of macroblock-level parallelization of H.264 decoding on a cc-NUMA multiprocessor architecture Mauricio Alvarez Alex Ramirez Mateo Valero Arnaldo Azevedo Cor Meenderinck Ben Juurlink Computación 264 synchronization multicore architectures Video codec parallelization multiprocessor architectures This paper presents a study of the performance scalability of a macroblocklevel parallelization of the H.264 decoder for High Definition (HD) applications on a multiprocessor architecture. We have implemented this parallelization on a cache coherent Nonuniform Memory Access (ccNUMA) shared memory multiprocessor (SMP) and compared the results with the theoretical expectations. The study includes the evaluation of three different scheduling techniques: static, dynamic and dynamic with tailsubmit. A dynamic scheduling approach with a tailsubmit optimization presents the best performance obtaining a maximum speedup of 9.5 with 24 processors. A detailed profiling analysis showed that thread synchronization is one of the limiting factors for achieving a better scalability. The paper includes an evaluation of the impact of using blocking synchronization APIs like POSIX threads and POSIX realtime extensions. Results showed that macroblocklevel parallelism as a very finegrain form of ThreadLevel Parallelism (TLP) is highly affected by the thread synchronization overhead generated by these APIs. Other synchronization methods, possibly with hardware support, are required in order to make MBlevel parallelization more scalable. 2009 artículo científico 1657-7663 https://www.redalyc.org/articulo.oa?id=133112608024 en http://www.redalyc.org/revista.oa?id=1331 Revista Avances en Sistemas e Informática application/pdf Universidad Nacional de Colombia Revista Avances en Sistemas e Informática (Colombia) Num.1 Vol.6 |
| format | Artículo científico |
| id | redalyc_133112608024 |
| language | en |
| publishDate | 2009 |
| publisher | Universidad Nacional de Colombia |
| spellingShingle | Performance evaluation of macroblock-level parallelization of H.264 decoding on a cc-NUMA multiprocessor architecture Mauricio Alvarez Computación 264 synchronization multicore architectures Video codec parallelization multiprocessor architectures Performance evaluation of macroblock-level parallelization of H.264 decoding on a cc-NUMA multiprocessor architecture Mauricio Alvarez Alex Ramirez Mateo Valero Arnaldo Azevedo Cor Meenderinck Ben Juurlink Computación 264 synchronization multicore architectures Video codec parallelization multiprocessor architectures This paper presents a study of the performance scalability of a macroblocklevel parallelization of the H.264 decoder for High Definition (HD) applications on a multiprocessor architecture. We have implemented this parallelization on a cache coherent Nonuniform Memory Access (ccNUMA) shared memory multiprocessor (SMP) and compared the results with the theoretical expectations. The study includes the evaluation of three different scheduling techniques: static, dynamic and dynamic with tailsubmit. A dynamic scheduling approach with a tailsubmit optimization presents the best performance obtaining a maximum speedup of 9.5 with 24 processors. A detailed profiling analysis showed that thread synchronization is one of the limiting factors for achieving a better scalability. The paper includes an evaluation of the impact of using blocking synchronization APIs like POSIX threads and POSIX realtime extensions. Results showed that macroblocklevel parallelism as a very finegrain form of ThreadLevel Parallelism (TLP) is highly affected by the thread synchronization overhead generated by these APIs. Other synchronization methods, possibly with hardware support, are required in order to make MBlevel parallelization more scalable. 2009 artículo científico 1657-7663 https://www.redalyc.org/articulo.oa?id=133112608024 en http://www.redalyc.org/revista.oa?id=1331 Revista Avances en Sistemas e Informática application/pdf Universidad Nacional de Colombia Revista Avances en Sistemas e Informática (Colombia) Num.1 Vol.6 |
| title | Performance evaluation of macroblock-level parallelization of H.264 decoding on a cc-NUMA multiprocessor architecture |
| topic | Computación 264 synchronization multicore architectures Video codec parallelization multiprocessor architectures |
| url | https://www.redalyc.org/articulo.oa?id=133112608024 |