HMMER performance model for multicore architectures

Sebastian Isaza, Ernst Houtgast, Georgi Gaydadjiev

Producción científica: Capítulo en Libro/ReporteContribución a la conferencia

1 Cita (Scopus)

Resumen

Exponential growth in biological sequence data combined with the computationally intensive nature of bioinformatics applications results in a continuously rising demand for processing power. In this paper, we propose a performance model that captures the behavior and performance scalability of HMMER, a bioinformatics application that identifies similarities between protein sequences and a protein family model. With our analytical model, the optimal master-worker ratio for a user scenario can be estimated. The model is evaluated and is found accurate with less than 2% error. We applied our model to a widely used heterogeneous multicore, the Cell BE, using the PPE and SPEs as master and workers respectively. Experimental results show that for the current parallelization strategy, the I/O speed at which the database is read from disk and the inputs pre-processing are the two most limiting factors in the Cell BE case.

Idioma originalInglés estadounidense
Título de la publicación alojadaProceedings - 2011 14th Euromicro Conference on Digital System Design
Subtítulo de la publicación alojadaArchitectures, Methods and Tools, DSD 2011
Páginas257-261
Número de páginas5
DOI
EstadoPublicada - 2011
Publicado de forma externa
Evento2011 14th Euromicro Conference on Digital System Design: Architectures, Methods and Tools, DSD 2011 - Oulu, Finlandia
Duración: ago. 31 2011sep. 2 2011

Serie de la publicación

NombreProceedings - 2011 14th Euromicro Conference on Digital System Design: Architectures, Methods and Tools, DSD 2011

Conferencia

Conferencia2011 14th Euromicro Conference on Digital System Design: Architectures, Methods and Tools, DSD 2011
País/TerritorioFinlandia
CiudadOulu
Período8/31/119/2/11

Áreas temáticas de ASJC Scopus

  • Hardware y arquitectura
  • Ingeniería de control y sistemas

Huella

Profundice en los temas de investigación de 'HMMER performance model for multicore architectures'. En conjunto forman una huella única.

Citar esto