A novel FPGA-based evolvable hardware system based on multiple processing arrays

In this paper, an architecture based on a scalable and flexible set of Evolvable Processing arrays is presented. FPGA-native Dynamic Partial Reconfiguration (DPR) is used for evolution, which is done intrinsically, letting the system to adapt autonomously to variable run-time conditions, including the presence of transient and permanent faults. The architecture supports different modes of operation, namely: independent, parallel, cascaded or bypass mode. These modes of operation can be used during evolution time or during normal operation. The evolvability of the architecture is combined with fault-tolerance techniques, to enhance the platform with self-healing features, making it suitable for applications which require both high adaptability and reliability. Experimental results show that such a system may benefit from accelerated evolution times, increased performance and improved dependability, mainly by increasing fault tolerance for transient and permanent faults, as well as providing some fault identification possibilities. The evolvable HW array shown is tailored for window-based image processing applications.

Submodular Optimization and Data Processing

Thesis (Ph.D.)--University of Washington, 2016-06

Investigation of Solid-State Shear Pulverization Processing Parameters for Polyethylene-Clay Nanocomposites

Investigates multiple processing parameters, includingpolymer type, filler type, processing technique, severity of SSSP (Solid-state shear pulverization)processing, and postprocessing, of SSSP. HDPE and LLDPE polymers with pristine clay and organo-clay samples are explored. Effects on crystallization, high-temperature behavior, mechanicalproperties, and gas barrier properties are examined. Thermal, mechanical, and morphological characterization is conducted to determine polymer/filler compatibility and superior processing methods for the polymer-clay nanocomposites.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks (WSNs) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of WSNs facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers (DCs). The high economical and environmental impact of the energy consumption in DCs requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of WSNs: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of DCs: energy-optimal workload assignment policies in heterogeneous DCs, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks ({WSNs}) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of {WSNs} facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers ({DCs}). The high economical and environmental impact of the energy consumption in {DCs} requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of {WSNs}: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of {DCs}: energy-optimal workload assignment policies in heterogeneous {DCs}, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

An architecture for reliable distributed computer-controlled systems

In Distributed Computer-Controlled Systems (DCCS), both real-time and reliability requirements are of major concern. Architectures for DCCS must be designed considering the integration of processing nodes and the underlying communication infrastructure. Such integration must be provided by appropriate software support services. In this paper, an architecture for DCCS is presented, its structure is outlined, and the services provided by the support software are presented. These are considered in order to guarantee the real-time and reliability requirements placed by current and future systems.

Time-domain optimization of amplifiers based on distributed genetic algorithms

Thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the subject of Electrical and Computer Engineering

A hybrid MAC Scheme to improve the transmission performance in body sensor networks

Wireless body sensor networks (WBSNs) constitute a key technology for closing the loop between patients and healthcare providers, as WBSNs provide sensing ability, as well as mobility and portability, essential characteristics for wide acceptance of wireless healthcare technology. However, one important and difficult aspect of WBSNs is to provide data transmissions with quality of service, among other factors due to the antennas being small size and placed close to the body. Such transmissions cannot be fully provided without the assumption of a MAC protocol that solves the problems of the medium sharing. A vast number of MAC protocols conceived for wireless networks are based on random or scheduled schemes. This paper studies firstly the suitability of two MAC protocols, one using CSMA and the other TDMA, to transmit directly to the base station the signals collected continuously from multiple sensor nodes placed on the human body. Tests in a real scenario show that the beaconed TDMA MAC protocol presents an average packet loss ratio lower than CSMA. However, the average packet loss ratio is above 1.0 %. To improve this performance, which is of vital importance in areas such as e-health and ambient assisted living, a hybrid TDMA/CSMA scheme is proposed and tested in a real scenario with two WBSNs and four sensor nodes per WBSN. An average packet loss ratio lower than 0.2 % was obtained with the hybrid scheme. To achieve this significant improvement, the hybrid scheme uses a lightweight algorithm to control dynamically the start of the superframes. Scalability and traffic rate variation tests show that this strategy allows approximately ten WBSNs operating simultaneously without significant performance degradation.

Membrane potential dynamics of neocortical projection neurons driving target-specific signals.

Primary sensory cortex discriminates incoming sensory information and generates multiple processing streams toward other cortical areas. However, the underlying cellular mechanisms remain unknown. Here, by making whole-cell recordings in primary somatosensory barrel cortex (S1) of behaving mice, we show that S1 neurons projecting to primary motor cortex (M1) and those projecting to secondary somatosensory cortex (S2) have distinct intrinsic membrane properties and exhibit markedly different membrane potential dynamics during behavior. Passive tactile stimulation evoked faster and larger postsynaptic potentials (PSPs) in M1-projecting neurons, rapidly driving phasic action potential firing, well-suited for stimulus detection. Repetitive active touch evoked strongly depressing PSPs and only transient firing in M1-projecting neurons. In contrast, PSP summation allowed S2-projecting neurons to robustly signal sensory information accumulated during repetitive touch, useful for encoding object features. Thus, target-specific transformation of sensory-evoked synaptic potentials by S1 projection neurons generates functionally distinct output signals for sensorimotor coordination and sensory perception.

El impacto de las aplicaciones intensivas de E/S en la planificación de trabajos en clusters no-dedicados

Con la mayor capacidad de los nodos de procesamiento en relación a la potencia de cómputo, cada vez más aplicaciones intensivas de datos como las aplicaciones de la bioinformática, se llevarán a ejecutar en clusters no dedicados. Los clusters no dedicados se caracterizan por su capacidad de combinar la ejecución de aplicaciones de usuarios locales con aplicaciones, científicas o comerciales, ejecutadas en paralelo. Saber qué efecto las aplicaciones con acceso intensivo a dados producen respecto a la mezcla de otro tipo (batch, interativa, SRT, etc) en los entornos no-dedicados permite el desarrollo de políticas de planificación más eficientes. Algunas de las aplicaciones intensivas de E/S se basan en el paradigma MapReduce donde los entornos que las utilizan, como Hadoop, se ocupan de la localidad de los datos, balanceo de carga de forma automática y trabajan con sistemas de archivos distribuidos. El rendimiento de Hadoop se puede mejorar sin aumentar los costos de hardware, al sintonizar varios parámetros de configuración claves para las especificaciones del cluster, para el tamaño de los datos de entrada y para el procesamiento complejo. La sincronización de estos parámetros de sincronización puede ser demasiado compleja para el usuario y/o administrador pero procura garantizar prestaciones más adecuadas. Este trabajo propone la evaluación del impacto de las aplicaciones intensivas de E/S en la planificación de trabajos en clusters no-dedicados bajo los paradigmas MPI y Mapreduce.

Scalability of efficient parallel K-Means

Clustering is defined as the grouping of similar items in a set, and is an important process within the field of data mining. As the amount of data for various applications continues to increase, in terms of its size and dimensionality, it is necessary to have efficient clustering methods. A popular clustering algorithm is K-Means, which adopts a greedy approach to produce a set of K-clusters with associated centres of mass, and uses a squared error distortion measure to determine convergence. Methods for improving the efficiency of K-Means have been largely explored in two main directions. The amount of computation can be significantly reduced by adopting a more efficient data structure, notably a multi-dimensional binary search tree (KD-Tree) to store either centroids or data points. A second direction is parallel processing, where data and computation loads are distributed over many processing nodes. However, little work has been done to provide a parallel formulation of the efficient sequential techniques based on KD-Trees. Such approaches are expected to have an irregular distribution of computation load and can suffer from load imbalance. This issue has so far limited the adoption of these efficient K-Means techniques in parallel computational environments. In this work, we provide a parallel formulation for the KD-Tree based K-Means algorithm and address its load balancing issues.

Dynamic load balancing in parallel KD-tree k-means

One among the most influential and popular data mining methods is the k-Means algorithm for cluster analysis. Techniques for improving the efficiency of k-Means have been largely explored in two main directions. The amount of computation can be significantly reduced by adopting geometrical constraints and an efficient data structure, notably a multidimensional binary search tree (KD-Tree). These techniques allow to reduce the number of distance computations the algorithm performs at each iteration. A second direction is parallel processing, where data and computation loads are distributed over many processing nodes. However, little work has been done to provide a parallel formulation of the efficient sequential techniques based on KD-Trees. Such approaches are expected to have an irregular distribution of computation load and can suffer from load imbalance. This issue has so far limited the adoption of these efficient k-Means variants in parallel computing environments. In this work, we provide a parallel formulation of the KD-Tree based k-Means algorithm for distributed memory systems and address its load balancing issue. Three solutions have been developed and tested. Two approaches are based on a static partitioning of the data set and a third solution incorporates a dynamic load balancing policy.

Non-uniform data distribution for communication-efficient parallel clustering

Global communicationrequirements andloadimbalanceof someparalleldataminingalgorithms arethe major obstacles to exploitthe computational power of large-scale systems. This work investigates how non-uniform data distributions can be exploited to remove the global communication requirement and to reduce the communication costin parallel data mining algorithms and, in particular, in the k-means algorithm for cluster analysis. In the straightforward parallel formulation of the k-means algorithm, data and computation loads are uniformly distributed over the processing nodes. This approach has excellent load balancing characteristics that may suggest it could scale up to large and extreme-scale parallel computing systems. However, at each iteration step the algorithm requires a global reduction operationwhichhinders thescalabilityoftheapproach.Thisworkstudiesadifferentparallelformulation of the algorithm where the requirement of global communication is removed, while maintaining the same deterministic nature ofthe centralised algorithm. The proposed approach exploits a non-uniform data distribution which can be either found in real-world distributed applications or can be induced by means ofmulti-dimensional binary searchtrees. The approachcanalso be extended to accommodate an approximation error which allows a further reduction ofthe communication costs. The effectiveness of the exact and approximate methods has been tested in a parallel computing system with 64 processors and in simulations with 1024 processing element

Perceptual Information Integration: Hypothetical Role of Astrocytes

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Design and implementation of wireless sensor networks for impedance-based structural health monitoring using ZigBee and Global System for Mobile Communications

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)