[Todos] Conferencia CANCELADA: Multimedia Workloads, Architectures, Performance and the use of Caches.

Esteban Feuerstein efeuerst en dc.uba.ar
Lun Dic 10 20:52:20 ART 2007


Por razones de fuerza mayor se suspendió el viaje del Prof. Smith a la
Argentina, por lo que la conferencia queda cancelada hasta nuevo aviso.
Lamento las molestias que pueda haber causado.
Esteban Feuerstein


  _____  

De: Esteban Feuerstein [mailto:efeuerst en dc.uba.ar] 
Enviado el: Lunes, 10 de Diciembre de 2007 03:44 p.m.
Para: 'Esteban Feuerstein'; 'docentes en dc.uba.ar'; 'alumnos en dc.uba.ar';
'medios en de.fcen.uba.ar'; 'dc-interna en dc.uba.ar'; 'redunci en info.unlp.edu.ar';
'sadio en speedy.org.ar'
CC: 'todos en df.uba.ar'; 'todos en dm.uba.ar'; 'alumnos en dm.uba.ar'
Asunto: Conferencia: Multimedia Workloads, Architectures, Performance and
the use of Caches.


El lunes 17 de diciembre próximo, en el Departamento de Computación dela
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,
Pabellón I de la Ciudad Universitaria, en horario y aula a designar, el
Prof.Alan Jay Smith, de la Computer Science Division de la Universidad de
Berkeley, dará una conferencia de título "Multimedia Workloads,
Architectures to Process Them, Their Performance and the Use of Caches for
Multimedia Workloads" 
La conferencia será en inglés.
 
Abstract:
The last decade has seen the integration of audio, video, and 3D graphics
into existing workloads as well as the emergence of new workloads dominated
by the processing of these forms of media.  Unfortunately, widely accepted
benchmarks which capture these new workloads in a realistic way have not
emerged. We present the Berkeley multimedia workload, which was developed to
facilitate our own studies on architectural support for multimedia.
 
The caching behavior of multimedia applications has been described as having
high instruction reference locality within small loops, very large working
sets, and poor data cache performance due to non-locality of data
references. Despite this, there is no published research deriving or
measuring these qualities. Utilizing the previously developed Berkeley
Multimedia Workload, we present the results of execution driven cache
simulations with the goal of aiding future media processing architecture
design. Our analysis examines the differences between multimedia and
traditional applications in cache behavior. We find that multimedia
applications actually exhibit lower instruction miss ratios and comparable
data miss ratios when contrasted with other widely studied workloads. In
addition, we find that longer data cache line sizes than are currently used
would benefit multimedia processing.
 
Many microprocessor instruction sets include instructions for accelerating
multimedia applications such as DVD playback, speech recognition and 3D
graphics. Despite general agreement on the need to support this emerging
workload, there are considerable differences between the instruction sets
that have been designed to do so.  We present a study of the performance of
five instruction sets on kernels extracted from a broad multimedia workload.
Each kernel was recoded in the assembly language of the five multimedia
extensions. We compare the performance of each extension against other
architectures as well as to the original compiled C performance. From our
analysis we determine how well multimedia workloads map to current
architectures, what was useful and what was not. We also propose two
enhancements to current architectures: strided memory operations, and
superwide registers.
 
The work to be presented was done with Nathan Slingerland, formerly a
graduate student at UC Berkeley.
 
 

------------ próxima parte ------------
Se ha borrado un adjunto en formato HTML...
URL: http://mail.df.uba.ar/pipermail/todos/attachments/20071210/5263bf7e/attachment.html 


Más información sobre la lista de distribución Todos