Pasar al contenido principal

MICROARQUITECTURA Y COMPILADORES PARA FUTUROS PROCESADORES III

TIN2013-44375-R

•
Nombre agencia financiadora Ministerio de Economía y Competitividad
Acrónimo agencia financiadora MINECO
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos I+D+i Retos de Investigación
Año convocatoria 2013
Plan Estatal 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Unidad de gestión Secretaría de Estado de Educación, Formación Profesional y Universidades
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA
Identificador persistente http://dx.doi.org/10.13039/501100003329

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

An ultra low-power hardware accelerator for acoustic scoring in speech recognition

  • Tabani, Hamid
  • Arnau Montañés, José María|||0000-0002-0336-9191
  • Tubella Murgadas, Jordi|||0000-0001-8669-0740
  • González Colás, Antonio María|||0000-0002-0009-0996
Accurate, real-time Automatic Speech Recognition (ASR) comes at a high energy cost, so accuracy has often to be sacrificed in order to fit the strict power constraints of mobile systems. However, accuracy is extremely important for the end-user, and today's systems are still unsatisfactory for many applications. The most critical component of an ASR system is the acoustic scoring, as it has a large impact on the accuracy of the system and takes up the bulk of execution time. The vast majority of ASR systems implement the acoustic scoring by means of Gaussian Mixture Models (GMMs), where the acoustic scores are obtained by evaluating multidimensional Gaussian distributions.In this paper, we propose a hardware accelerator for GMM evaluation that reduces the energy required for acoustic scoring by three orders of magnitude compared to solutions based on CPUs and GPUs. Our accelerator implements a lazy evaluation scheme where Gaussians are computed on demand, avoiding 50% of the computations. Furthermore, it employs a novel clustering scheme to reduce the size of the acoustic model, which results in 8x memory bandwidth savings with a negligible impact on accuracy. Finally, it includes a novel memoization scheme that avoids 74.88% of floating-point operations. The end design provides a 164x speedup and 3532x energy reduction when compared with a highly-tuned implementation running on a modern mobile CPU. Compared to a state-of-the-art mobile GPU, the GMM accelerator achieves 5.89x speedup over a highly optimized CUDA implementation, while reducing energy by 241x., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

A novel register renaming technique for out-of-order processors

  • Tabani, Hamid
  • Arnau Montañés, José María|||0000-0002-0336-9191
  • Tubella Murgadas, Jordi|||0000-0001-8669-0740
  • González Colás, Antonio María|||0000-0002-0009-0996
Modern superscalar processors support a large number of in-flight instructions, which requires sizeable register files. Conventional register renaming techniques allocate a new storage location, i.e. physical register, for every instruction whose destination is a logical register in order to remove false dependences. Physical registers are released in a conservative manner when the same logical register is redefined. For this reason, many cycles may happen between the last read and the release of a physical register, leading to suboptimal utilization of the register file. We have observed that for more than 50% of the instructions in SPECfp and more than 30% of the instructions in SPECint that have a destination register, the produced value has only a single consumer. In this case, the RAW dependence guarantees that the producer-consumer instructions pair will be executed in program order and, hence, the same physical register can be used to store the value produced by both instructions. In this paper, we propose a renaming technique that exploits this property to reduce the pressure on the register file. Our technique leverages physical register sharing by introducing minor changes in the register map table and the issue queue. We also describe how our renaming scheme supports precise exceptions. We evaluated our renaming technique on top of a modern out-of-order processor. Our experimental results show that it provides 6% speedup on average for the SPEC2006 benchmarks. Alternatively, our renaming scheme achieves the same performance while reducing the number of physical registers by 10.5%., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

A detailed methodology to compute soft error rates in advanced technologies

  • Riera Villanueva, Marc|||0000-0002-2768-5703
  • Canal Corretger, Ramon|||0000-0003-4542-204X
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
  • González Colás, Antonio María|||0000-0002-0009-0996
System reliability has become a key design aspect for computer systems due to the aggressive technology miniaturization. Errors are typically dominated by transient faults due to radiation and are strongly related to the technology used to build hardware. However, there is a lack of detailed methodologies to model and fairly compare Soft Error Rates (SER) across different advanced technologies. This work first describes a common methodology that from (1) technology models, (2) location (latitude, longitude and altitude), (3) operating conditions and (4) circuit descriptions (i.e. SRAM, latches, logic gates) can obtain accurate Soft Error Rates. Then, we use it to characterize soft errors through current and future technologies. Results at the technology layer show that new technologies, such as FinFET and SOI, can reduce SER up to 100x while the location can increase SER up to 650x. © 2016 EDAA., This work has been partially supported by the Spanish Ministry of Education and Science under grant TIN2013-44375-R and the FP7 program of the EU under contract FP7-611404 (CLERECO). Jaume Abella has been partially supported by the Ministry of Economy and Competitiveness under Ramon y Cajal postdoctoral fellowship number RYC-2013-14717., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

A novel register renaming technique for out-of-order processors

  • Tabani, Hamid
  • Arnau Montañés, José María
  • Tubella Murgadas, Jordi
  • González Colás, Antonio María
Modern superscalar processors support a large number of in-flight instructions, which requires sizeable register files. Conventional register renaming techniques allocate a new storage location, i.e. physical register, for every instruction whose destination is a logical register in order to remove false dependences. Physical registers are released in a conservative manner when the same logical register is redefined. For this reason, many cycles may happen between the last read and the release of a physical register, leading to suboptimal utilization of the register file. We have observed that for more than 50% of the instructions in SPECfp and more than 30% of the instructions in SPECint that have a destination register, the produced value has only a single consumer. In this case, the RAW dependence guarantees that the producer-consumer instructions pair will be executed in program order and, hence, the same physical register can be used to store the value produced by both instructions. In this paper, we propose a renaming technique that exploits this property to reduce the pressure on the register file. Our technique leverages physical register sharing by introducing minor changes in the register map table and the issue queue. We also describe how our renaming scheme supports precise exceptions. We evaluated our renaming technique on top of a modern out-of-order processor. Our experimental results show that it provides 6% speedup on average for the SPEC2006 benchmarks. Alternatively, our renaming scheme achieves the same performance while reducing the number of physical registers by 10.5%., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

A detailed methodology to compute soft error rates in advanced technologies

  • Riera Villanueva, Marc
  • Canal Corretger, Ramon
  • Abella Ferrer, Jaume
  • González Colás, Antonio María
System reliability has become a key design aspect for computer systems due to the aggressive technology miniaturization. Errors are typically dominated by transient faults due to radiation and are strongly related to the technology used to build hardware. However, there is a lack of detailed methodologies to model and fairly compare Soft Error Rates (SER) across different advanced technologies. This work first describes a common methodology that from (1) technology models, (2) location (latitude, longitude and altitude), (3) operating conditions and (4) circuit descriptions (i.e. SRAM, latches, logic gates) can obtain accurate Soft Error Rates. Then, we use it to characterize soft errors through current and future technologies. Results at the technology layer show that new technologies, such as FinFET and SOI, can reduce SER up to 100x while the location can increase SER up to 650x. © 2016 EDAA., This work has been partially supported by the Spanish Ministry of Education and Science under grant TIN2013-44375-R and the FP7 program of the EU under contract FP7-611404 (CLERECO). Jaume Abella has been partially supported by the Ministry of Economy and Competitiveness under Ramon y Cajal postdoctoral fellowship number RYC-2013-14717., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

An ultra low-power hardware accelerator for acoustic scoring in speech recognition

  • Tabani, Hamid
  • Arnau Montañés, José María
  • Tubella Murgadas, Jordi
  • González Colás, Antonio María
Accurate, real-time Automatic Speech Recognition (ASR) comes at a high energy cost, so accuracy has often to be sacrificed in order to fit the strict power constraints of mobile systems. However, accuracy is extremely important for the end-user, and today's systems are still unsatisfactory for many applications. The most critical component of an ASR system is the acoustic scoring, as it has a large impact on the accuracy of the system and takes up the bulk of execution time. The vast majority of ASR systems implement the acoustic scoring by means of Gaussian Mixture Models (GMMs), where the acoustic scores are obtained by evaluating multidimensional Gaussian distributions.In this paper, we propose a hardware accelerator for GMM evaluation that reduces the energy required for acoustic scoring by three orders of magnitude compared to solutions based on CPUs and GPUs. Our accelerator implements a lazy evaluation scheme where Gaussians are computed on demand, avoiding 50% of the computations. Furthermore, it employs a novel clustering scheme to reduce the size of the acoustic model, which results in 8x memory bandwidth savings with a negligible impact on accuracy. Finally, it includes a novel memoization scheme that avoids 74.88% of floating-point operations. The end design provides a 164x speedup and 3532x energy reduction when compared with a highly-tuned implementation running on a modern mobile CPU. Compared to a state-of-the-art mobile GPU, the GMM accelerator achieves 5.89x speedup over a highly optimized CUDA implementation, while reducing energy by 241x., Peer Reviewed, Postprint (published version)