%0 Journal Article
%T Dynamic Reconfigurable Computing: The Alternative to Homogeneous Multicores under Massive Defect Rates
%A Monica Magalh£¿es Pereira
%A Luigi Carro
%J International Journal of Reconfigurable Computing
%D 2011
%I Hindawi Publishing Corporation
%R 10.1155/2011/452589
%X The aggressive scaling of CMOS technology has increased the density and allowed the integration of multiple processors into a single chip. Although solutions based on MPSoC architectures can increase application's speed through TLP exploitation, this speedup is still limited to the amount of parallelism available in the application, as demonstrated by Amdahl's Law. Moreover, with the continuous shrinking of device features, very aggressive defect rates are expected for new technologies. Under high defect rates a large amount of processors of the MPSoC will be susceptible to defects and consequently will fail, not only reducing yield but also severely affecting the expected performance. This paper presents a run-time adaptive architecture that allows software execution even under aggressive defect rates. The proposed architecture can accelerate not only highly parallel applications but also sequential ones, and it is a heterogeneous solution to overcome the performance penalty that is imposed to homogeneous MPSoCs under massive defect rates. 1. Introduction The scaling of CMOS technology has increased the density and consequently made the integration of several processors in one chip possible. Many architectural solutions with several cores can be found in the literature in the past decade [1]. These solutions are mainly used to accelerate execution through task level parallelism (TLP) exploitation. However, the speedup achieved by these systems is limited to the amount of parallelism available in the applications, as already foreseen by Amdahl [2]. According to Rutzig et al. [3], current embedded system domain applications present a heterogeneous behavior that includes not only highly parallel applications but also general purpose applications that are also migrating to embedded system domain, such as browsers and high definition video processing. To cope with this heterogeneous behavior an architecture design is necessary to not only accelerate execution through TLP but also find other ways to accelerate software execution, for example, through ILP (instruction level parallelism) exploitation, or accelerate sequential execution. One solution to overcome Amdahl¡¯s law and sustain the speedup of MPSoCs is the use of heterogeneous cores, where each core is specialized in different application sets. An example of a heterogeneous architecture is the Samsung S5PC100 [4] used in Apple¡¯s iPhone technology [5]. The Samsung S5PC100 is a multimedia-based MPSoC that has an ARM-based central general purpose processor and five multimedia accelerators targeted to DSP
%U http://www.hindawi.com/journals/ijrc/2011/452589/