Les composants reprogrammables de type FPGA à base de mémoire SRAM sont des candidats appréciés pour les applications aéronautiques et spatiales. Cependant les particules énergétiques présentes dans l'environnement naturel peuvent engendrer une mutation de l'application implémentée en créant des erreurs dans la mémoire de configuration. Les travaux réalisés au cours de cette thèse ont eu pour but principal l'étude d'une stratégie de prédiction du taux d'erreurs pour un système implémenté dans ce type de composant. La pertinence d'une telle approche a été évaluée par confrontation des prédictions des taux d'erreurs, issus de sessions d'injections matérielles/logicielles de fautes, avec les mesures obtenues lors de campagnes de test en accélérateur de particules. Le second objectif fut le développement d'une expérience embarquée, dans un satellite scientifique de la NASA, afin d'obtenir des informations sur le comportement du FPGA étudié et son application en environnement réel.

The power consumption in commercial processors and application specific integrated circuits increases with decreasing technology nodes. Power saving techniques have become a first class design point for current and future VLSI systems. These systems employ large on-chip SRAM memories. Reducing memory leakage power while maintaining data integrity is a key criterion for modern day systems. Unfortunately, state of the art techniques like power-gating can only be applied to logic as these would destroy the contents of the memory if applied to a SRAM system. Fortunately, previous works have noted large temporal and spatial locality for data patterns in commerical processors as well as application specific ICs that work on images, audio and video data. This thesis presents a novel column based Energy Compression technique that saves SRAM power by selectively turning off cells based on a data pattern. This technique is applied to study the power savings in application specific inegrated circuit SRAM memories and can also be applied for commercial processors. The thesis also evaluates the effects of processing images before storage and data cluster patterns for optimizing power savings.

Aggressive scaling of CMOS technology in sub-90nm nodes has created huge challenges. Variations due to fundamental physical limits, such as random dopants fluctuation (RDF) and line edge roughness (LER) are increasing significantly with technology scaling. In addition, manufacturing tolerances in process technology are not scaling at the same pace as transistor's channel length due to process control limitations (e.g., sub-wavelength lithography). Therefore, within-die process variations worsen with successive technology generations. These variations have a strong impact on the maximum clock frequency and leakage power for any digital circuit, and can also result in functional yield losses in variation-sensitive digital circuits (such as SRAM). Moreover, in nanometer technologies, digital circuits show an increased sensitivity to process variations due to low-voltage operation requirements, which are aggravated by the strong demand for lower power consumption and cost while achieving higher performance and density. It is therefore not surprising that the International Technology Roadmap for Semiconductors (ITRS) lists variability as one of the most challenging obstacles for IC design in nanometer regime. To facilitate variation-tolerant design, we study the impact of random variations on the delay variability of a logic gate and derive simple and scalable statistical models to evaluate delay variations in the presence of within-die variations. This work provides new design insight and highlights the importance of accounting for the effect of input slew on delay variations, especially at lower supply voltages. The derived models are simple, scalable, bias dependent and only require the knowledge of easily measurable parameters. This makes them useful in early design exploration, circuit/architecture optimization as well as technology prediction (especially in low-power and low-voltage operation). The derived models are verified using Monte Carlo SPICE simulations using industrial 90nm technology. Random variations in nanometer technologies are considered one of the largest design considerations. This is especially true for SRAM, due to the large variations in bitcell characteristics. Typically, SRAM bitcells have the smallest device sizes on a chip. Therefore, they show the largest sensitivity to different sources of variations. With the drastic increase in memory densities, lower supply voltages and higher variations, statistical simulation methodologies become imperative to estimate memory yield and optimize performance and power. In this research, we present a methodology for statistical simulation of SRAM read access yield, which is tightly related to SRAM performance and power consumption. The proposed flow accounts for the impact of bitcell read current variation, sense amplifier offset distribution, timing window variation and leakage variation on functional yield. The methodology overcomes the pessimism existing in conventional worst-case design techniques that are used in SRAM design. The proposed statistical yield estimation methodology allows early yield prediction in the design cycle, which can be used to trade off performance and power requirements for SRAM. The methodology is verified using measured silicon yield data from a 1Mb memory fabricated in an industrial 45nm technology. Embedded SRAM dominates modern SoCs and there is a strong demand for SRAM with lower power consumption while achieving high performance and high density. However, in the presence of large process variations, SRAMs are expected to consume larger power to ensure correct read operation and meet yield targets. We propose a new architecture that significantly reduces array switching power for SRAM. The proposed architecture combines built-in self-test (BIST) and digitally controlled delay elements to reduce the wordline pulse width for memories while ensuring correct read operation; hence, reducing switching power. A new statistical simulation flow was developed to evaluate the power savings for the proposed architecture. Monte Carlo simulations using a 1Mb SRAM macro from an industrial 45nm technology was used to examine the power reduction achieved by the system. The proposed architecture can reduce the array switching power significantly and shows large power saving - especially as the chip level memory density increases. For a 48Mb memory density, a 27% reduction in array switching power can be achieved for a read access yield target of 95%. In addition, the proposed system can provide larger power saving as process variations increase, which makes it a very attractive solution for 45nm and below technologies. In addition to its impact on bitcell read current, the increase of local variations in nanometer technologies strongly affect SRAM cell stability. In this research, we propose a novel single supply voltage read assist technique to improve SRAM static noise margin (SNM). The proposed technique allows precharging different parts of the bitlines to VDD and GND and uses charge sharing to precisely control the bitline voltage, which improves the bitcell stability. In addition to improving SNM, the proposed technique also reduces memory access time. Moreover, it only requires one supply voltage, hence, eliminates the need of large area voltage shifters. The proposed technique has been implemented in the design of a 512kb memory fabricated in 45nm technology. Results show improvements in SNM and read operation window which confirms the effectiveness and robustness of this technique.

Aggressive scaling of CMOS technology in sub-90nm nodes has created huge challenges. Variations due to fundamental physical limits, such as random dopants fluctuation (RDF) and line edge roughness (LER) are increasing significantly with technology scaling. In addition, manufacturing tolerances in process technology are not scaling at the same pace as transistor's channel length due to process control limitations (e.g., sub-wavelength lithography). Therefore, within-die process variations worsen with successive technology generations. These variations have a strong impact on the maximum clock frequency and leakage power for any digital circuit, and can also result in functional yield losses in variation-sensitive digital circuits (such as SRAM). Moreover, in nanometer technologies, digital circuits show an increased sensitivity to process variations due to low-voltage operation requirements, which are aggravated by the strong demand for lower power consumption and cost while achieving higher performance and density. It is therefore not surprising that the International Technology Roadmap for Semiconductors (ITRS) lists variability as one of the most challenging obstacles for IC design in nanometer regime. To facilitate variation-tolerant design, we study the impact of random variations on the delay variability of a logic gate and derive simple and scalable statistical models to evaluate delay variations in the presence of within-die variations. This work provides new design insight and highlights the importance of accounting for the effect of input slew on delay variations, especially at lower supply voltages. The derived models are simple, scalable, bias dependent and only require the knowledge of easily measurable parameters. This makes them useful in early design exploration, circuit/architecture optimization as well as technology prediction (especially in low-power and low-voltage operation). The derived models are verified using Monte Carlo SPICE simulations using industrial 90nm technology. Random variations in nanometer technologies are considered one of the largest design considerations. This is especially true for SRAM, due to the large variations in bitcell characteristics. Typically, SRAM bitcells have the smallest device sizes on a chip. Therefore, they show the largest sensitivity to different sources of variations. With the drastic increase in memory densities, lower supply voltages and higher variations, statistical simulation methodologies become imperative to estimate memory yield and optimize performance and power. In this research, we present a methodology for statistical simulation of SRAM read access yield, which is tightly related to SRAM performance and power consumption. The proposed flow accounts for the impact of bitcell read current variation, sense amplifier offset distribution, timing window variation and leakage variation on functional yield. The methodology overcomes the pessimism existing in conventional worst-case design techniques that are used in SRAM design. The proposed statistical yield estimation methodology allows early yield prediction in the design cycle, which can be used to trade off performance and power requirements for SRAM. The methodology is verified using measured silicon yield data from a 1Mb memory fabricated in an industrial 45nm technology. Embedded SRAM dominates modern SoCs and there is a strong demand for SRAM with lower power consumption while achieving high performance and high density. However, in the presence of large process variations, SRAMs are expected to consume larger power to ensure correct read operation and meet yield targets. We propose a new architecture that significantly reduces array switching power for SRAM. The proposed architecture combines built-in self-test (BIST) and digitally controlled delay elements to reduce the wordline pulse width for memories while ensuring correct read operation; hence, reducing switching power. A new statistical simulation flow was developed to evaluate the power savings for the proposed architecture. Monte Carlo simulations using a 1Mb SRAM macro from an industrial 45nm technology was used to examine the power reduction achieved by the system. The proposed architecture can reduce the array switching power significantly and shows large power saving - especially as the chip level memory density increases. For a 48Mb memory density, a 27% reduction in array switching power can be achieved for a read access yield target of 95%. In addition, the proposed system can provide larger power saving as process variations increase, which makes it a very attractive solution for 45nm and below technologies. In addition to its impact on bitcell read current, the increase of local variations in nanometer technologies strongly affect SRAM cell stability. In this research, we propose a novel single supply voltage read assist technique to improve SRAM static noise margin (SNM). The proposed technique allows precharging different parts of the bitlines to VDD and GND and uses charge sharing to precisely control the bitline voltage, which improves the bitcell stability. In addition to improving SNM, the proposed technique also reduces memory access time. Moreover, it only requires one supply voltage, hence, eliminates the need of large area voltage shifters. The proposed technique has been implemented in the design of a 512kb memory fabricated in 45nm technology. Results show improvements in SNM and read operation window which confirms the effectiveness and robustness of this technique.

Technology scaling has been the most obvious choice of designers and chip manufacturing companies to improve the performance of analog and digital circuits. With the ever shrinking technological node, process variations can no longer be ignored and play a significant role in determining the performance of nanoscaled devices. By choosing a worst case design methodology, circuit designers have been very munificent with the design parameters chosen, often manifesting in pessimistic designs with significant area overheads. Significant work has been done in estimating the impact of intra-die process variations on circuit performance, pertinently, noise margin and standby leakage power, for fixed transistor channel dimensions. However, for an optimal, high yield, SRAM cell design, it is absolutely imperative to analyze the impact of process variations at every design point, especially, since the distribution of process variations is a statistically varying parameter and has an inverse correlation with the area of the MOS transistor. Furthermore, the first order analytical models used for optimization of SRAM memories are not as accurate and the impact of voltage and its inclusion as an input, along with other design parameters, is often ignored. In this thesis, the performance parameters of a nano-scaled 6-T SRAM cell are modeled as an accurate, yield aware, empirical polynomial predictor, in the presence of intra-die process variations. The estimated empirical models are used in a constrained non-linear, robust optimization framework to design an SRAM cell, for a 45 nm CMOS technology, having optimal performance, according to bounds specified for the circuit performance parameters, with the objective of minimizing on-chip area. This statistically aware technique provides a more realistic design methodology to study the trade off between performance parameters of the SRAM. Furthermore, a dual optimization approach is followed by considering SRAM power supply and wordline voltages as additional input parameters, to simultaneously tune the design parameters, ensuring a high yield and considerable area reduction. In addition, the cell level optimization framework is extended to the system level optimization of caches, under both cell level and system level performance constraints.

With a view to reduce the number of components in a Static Random Access Memory (SRAM) cell, the feasibility of use of Lambda Bipolar Transistor (LBT)in the bistable element of the cell has been explored under the present study. The LBT under consideration here comprises of an enhancement mode MOSFET integrated with a parasitic bipolar transistor so as to perform as a negative resistance device. LBTs for the study have been fabricated and analysed. The devices have been shown to function at much lower voltage and current levels than those reported earlier/ and thus have been shown to be suitable for lower power applications. The issues of agreements and discrepancies of the experimental results with the original DC model of the device have been highlighted and discussed. The factors contributing to the drain current of the MOSFET in the LBT have been identified. It has also been shown that in the real case of an LBT in operation, the MOSFET in it does not function as a discrete device for the same conditions of voltages and current levels as in an LBT. As per the present study, it is assessed to be influenced by the presence of the BJT in operation and this effect is felt more at the lower current levels of operation. With a separate and tailored p-well implantation the possibility of fabrication of LBTs with a CMOS technology is established. Along with a couple of polysilicon resistors, the LBTs have been successfully made to perform in the common-collector configuration as the bistable storage element of SRAM cell (as proposed in the literature). The bistable element with the LBT in common-emitter mode also has been visualised and practically achieved with the fabricated devices. The WRITE transients for either case have been simulated for various levels of WRITE voltages and their time of hold.The speed of Writing achieved are found comparable with that of the standard SRAMs. The advantages and disadvantages of using the LBT in either mode have been highlighted and discussed. The power consumption of the bistable element with the LBT in either mode is however shown to be the same. A different approach of READING has been proposed to overcome the factors known to increase the cycle time. On the whole, under the present study, the proposal of using LBTs in the bistable storage element of the SRAM cell has been shown to be feasible. Such SRAM circuits can find possible applications in the fields where smaller circuit area is the major concern.

Parallèlement à l'accroissement de la part dévolue à la mémoire au sein des circuits, l'évolution technologique s'accompagne d'uneaugmentation de la variabilité des performances, notamment dues aux variations de procédés de fabrication, de la tension d'alimentation et de la température. En terme de conception, ces variations de procédés de fabrication et de conditions de fonctionnement sont généralement prises en compte en adoptant une approche pire et meilleur cas(méthode des corners). Cependant, l'accroissement de la variabilité des procédés de fabrication conduit à l'accroissement relatif de la fourchette d'estimation des performances, comme la marge temporelle de lecture dans une mémoire embarquée de type SRAM. Ainsi la seule alternative, permettant de s'affranchir de la méthode des corners, réside dans l'adoption de techniques statistiqueset notamment l'analyse statistique des performances temporelle. Cette analyse statistique des performances des SRAMs constitue le cœur de cette thèse. Dans un premier temps, nous avons démontré les limites de la méthode des corners sur la marge temporelle de lecture de la SRAM. Puis, nous avons développé une approche de modélisation permettant de s'affranchir de la méthode des corners. Cette approche a ensuite été utilisée dans le dimensionnement statistique de la mémoire afin d'optimiser ses performances temporelles, réduisant ainsi lamarge excessive de lecture introduite par l'approche traditionnelle.

Cette thèse s'intéresse en premier lieu à l'évaluation des effets fonctionnels des erreurs survenant dans la mémoire SRAM de configuration de certains FPGAs. La famille Virtex II Pro de Xilinx est utilisée comme premier cas pratique d'expérimentation. Des expérimentations sous faisceau laser nous ont permis d'avoir une bonne vue d'ensemble sur les motifs d'erreurs réalistes qui sont obtenus par des sources de perturbations réelles. Une méthodologie adaptée d'injection de fautes a donc été définie pour permettre une meilleure évaluation, en phase de conception, de la robustesse d'un circuit implanté sur ce type de technologie. Cette méthodologie est basée sur de la reconfiguration dynamique. Le même type d'approche a ensuite été évalué sur plusieurs cibles technologiques, ce qui a nécessité le développement de plusieurs environnements d'injection de fautes. L'étude a pour la première fois inclus la famille AT40K de ATMEL, qui permet un type de reconfiguration unique et efficace. Le second type de contribution concerne l'augmentation à faible coût de la robustesse de circuits implantés sur des plateformes FPGA SRAM. Nous proposons une approche de protection sélective exploitant les ressources du FPGA inutilisées par l'application. L'approche a été automatisée sur plusieurs cibles technologiques (Xilinx, Altera) et l'efficacité est analysée en utilisant les méthodes d'injection de fautes précédemment développées.

Nesta dissertação é apresentado o estudo da célula SRAM estática de 6 transistores, com tecnologia CMOS, sendo utilizada em ambiente exposto à radiação. Foi verificado, através de simulação com o Hspice (HSPICE, 2009; KIME, 1998) e com a análise de Monte Carlo, o seu comportamento com relação à dose de ionização total (Total Ionization Dose, TID), a qual altera a tensão de limiar (threshold voltage, Vth) e a corrente de fuga, não sendo utilizada nenhuma técnica de fabricação especial para tolerância à radiação. Na simulação foi observado o comportamento da célula com relação ao tempo de atraso de escrita, à margem de ruído de leitura e ao consumo de energia. As simulações incluem as tecnologias de 130nm e 350nm sendo, portanto, possível comparar os efeitos de radiação citados em ambas, para verificar qual é a mais naturalmente resistente a radiação, verificando se está coerente com resultados divulgados na literatura. Para simular o efeito de dose, altera-se a tensão de limiar (threshold voltage, Vth) com a análise de Monte Carlo e, para a corrente de fuga, adiciona-se uma fonte de corrente entre o dreno e fonte de cada transistor. Os valores de Vth e corrente de fuga foram obtidos nas referências (HAUGERUD, 2005) para a tecnologia 130nm e (LACOE, 1998) para a tecnologia 350 nm. As simulações mostram que o comportamento foi coerente com resultados já conhecidos, em que a tecnologia mais antiga (350nm) tem alterações mais significativas do que a tecnologia mais atual, em relação à TID. / This work presents the study of the static RAM (SRAM) cell with 6 transistor, using CMOS technology, under radiation environment. The electrical behavior of the cell is evaluated using SPICE simulation (HSPICE, 2009; KIME, 1998) and applying Monte Carlo analysis. The effect of total ionization dose is analyzed through the modeling of threshold voltage shifts and leakage currents. The case study processes of this work do not use any special fabrication steps to make the circuit tolerant to radiation. The behavior of the cell related to write propagation time, read noise margin and energy consumption is evaluated through scripts written to support the simulation campaign. The simulations were performed for both 130nm and 350nm technologies, making possible to compare which one is more resistant to radiation. To further explore the dose effect in the case where the radiation does not affect all transistors in exactly the same way, the threshold voltage (Vth) of the transistors is varied randomly in the Monte Carlo analysis. To consider the leakage current, it is added a current source between drain and source of each transistor. The values of Vth and leakage current were obtained in reference (HAUGERUD, 2005) for the 130nm and in reference (LACOE, 1998) for the 350nm technology. The simulations show that the behavior was consistent with results already known, in which the older technology (350nm) is more significant changes then the most current technology, for the TID.

Confiabilidade é um parâmetro de projeto importante para aplicações criticas tanto na Terra como também no espaço. Os FPGAs baseados em memoria SRAM são atrativos para implementar aplicações criticas devido a seu alto desempenho e flexibilidade. No entanto, estes FPGAs são susceptíveis aos efeitos da radiação tais como os erros transientes na memoria de configuração. Além disso, outros efeitos como o envelhecimento (aging) ou escalonamento da tensão de alimentação (voltage scaling) incrementam a sensibilidade à radiação dos FPGAs. Nossos resultados experimentais mostram que o envelhecimento e o escalonamento da tensão de alimentação podem aumentar ao menos duas vezes a susceptibilidade de FPGAs baseados em SRAM a erros transientes. Estes resultados são inovadores porque estes combinam três efeitos reais que acontecem em FPGAs baseados em SRAM. Os resultados podem guiar aos projetistas a prever os efeitos dos erros transientes durante o tempo de operação do dispositivo em diferentes níveis de tensão. A correção da memoria usando a técnica de scrubbing é um método efetivo para corrigir erros transientes em memorias SRAM, mas este método impõe custos adicionais em termos de área e consumo de energia. Neste trabalho, nos propomos uma nova técnica de scrubbing usando a redundância interna a nível de quadros chamada FLR- scrubbing. Esta técnica possui mínimo consumo de energia sem comprometer a capacidade de correção. Como estudo de caso, a técnica foi implementada em um FPGA de tamanho médio Xilinx Virtex-5, ocupando 8% dos recursos disponíveis e consumindo seis vezes menos energia que um circuito corretor tradicional chamado blind scrubber. Além, a técnica proposta reduz o tempo de reparação porque evita o uso de uma memoria externa como referencia. E como outra contribuição deste trabalho, nos apresentamos os detalhes de uma plataforma de injeção de falhas múltiplas que permite emular os erros transientes na memoria de configuração do FPGA usando reconfiguração parcial dinâmica. Resultados de campanhas de injeção são apresentados e comparados com experimentos de radiação acelerada. Finalmente, usando a plataforma de injeção de falhas proposta, nos conseguimos analisar a efetividade da técnica FLR-scrubbing. Nos também confirmamos estes resultados com experimentos de radiação acelerada. / Reliability is an important design constraint for critical applications at ground-level and aerospace. SRAM-based FPGAs are attractive for critical applications due to their high performance and flexibility. However, they are susceptible to radiation effects such as soft errors in the configuration memory. Furthermore, the effects of aging and voltage scaling increment the sensitivity of SRAM-based FPGAs to soft errors. Experimental results show that aging and voltage scaling can increase at least two times the susceptibility of SRAM-based FPGAs to Soft Error Rate (SER). These findings are innovative because they combine three real effects that occur in SRAM-based FPGAs. Results can guide designers to predict soft error effects during the lifetime of devices operating at different power supply voltages. Memory scrubbing is an effective method to correct soft errors in SRAM memories, but it imposes an overhead in terms of silicon area and energy consumption. In this work, it is proposed a novel scrubbing technique using internal frame redundancy called Frame-level Redundancy Scrubbing (FLRscrubbing) with minimum energy consumption overhead without compromising the correction capabilities. As a case study, the FLR-scrubbing controller was implemented on a mid-size Xilinx Virtex-5 FPGA device, occupying 8% of available slices and consumes six times less energy per scrubbed frame than a classic blind scrubber. Also, the technique reduces the repair time by avoiding the use of an external golden memory for reference. As another contribution, this work presents the details of a Multiple Fault Injection Platform that emulates the configuration memory upsets of an FPGA using dynamic partial reconfiguration. Results of fault injection campaigns are presented and compared with accelerated ground-level radiation experiments. Finally, using our proposed fault injection platform it was possible to analyze the effectiveness of the FLR-scrubbing technique. Accelerated radiation tests confirmed these results.