Makespan computation for GPU threads running on a single streaming multiprocessor

Graphics processors were originally developed for rendering graphics but have recently evolved towards being an architecture for general-purpose computations. They are also expected to become important parts of embedded systems hardware -- not just for graphics. However, this necessitates the development of appropriate timing analysis techniques which would be required because techniques developed for CPU scheduling are not applicable. The reason is that we are not interested in how long it takes for any given GPU thread to complete, but rather how long it takes for all of them to complete. We therefore develop a simple method for finding an upper bound on the makespan of a group of GPU threads executing the same program and competing for the resources of a single streaming multiprocessor (whose architecture is based on NVIDIA Fermi, with some simplifying assunptions). We then build upon this method to formulate the derivation of the exact worst-case makespan (and corresponding schedule) as an optimization problem. Addressing the issue of tractability, we also present a technique for efficiently computing a safe estimate of the worstcase makespan with minimal pessimism, which may be used when finding an exact value would take too long.

Building a microscope for the data center

Managing the physical and compute infrastructure of a large data center is an embodiment of a Cyber-Physical System (CPS). The physical parameters of the data center (such as power, temperature, pressure, humidity) are tightly coupled with computations, even more so in upcoming data centers, where the location of workloads can vary substantially due, for example, to workloads being moved in a cloud infrastructure hosted in the data center. In this paper, we describe a data collection and distribution architecture that enables gathering physical parameters of a large data center at a very high temporal and spatial resolutionof the sensor measurements. We think this is an important characteristic to enable more accurate heat-flow models of the data center andwith them, _and opportunities to optimize energy consumption. Havinga high resolution picture of the data center conditions, also enables minimizing local hotspots, perform more accurate predictive maintenance (pending failures in cooling and other infrastructure equipment can be more promptly detected) and more accurate billing. We detail this architecture and define the structure of the underlying messaging system that is used to collect and distribute the data. Finally, we show the results of a preliminary study of a typical data center radio environment.

A framework for offloading real-time applications in a distributed environment

This work focuses on highly dynamic distributed systems with Quality of Service (QoS) constraints (most importantly real-time constraints). To that purpose, real-time applications may benefit from code offloading techniques, so that parts of the application can be offloaded and executed, as services, by neighbour nodes, which are willing to cooperate in such computations. These applications explicitly state their QoS requirements, which are translated into resource requirements, in order to evaluate the feasibility of accepting other applications in the system.

Pattern operators for grid

The definition and programming of distributed applications has become a major research issue due to the increasing availability of (large scale) distributed platforms and the requirements posed by the economical globalization. However, such a task requires a huge effort due to the complexity of the distributed environments: large amount of users may communicate and share information across different authority domains; moreover, the “execution environment” or “computations” are dynamic since the number of users and the computational infrastructure change in time. Grid environments, in particular, promise to be an answer to deal with such complexity, by providing high performance execution support to large amount of users, and resource sharing across different organizations. Nevertheless, programming in Grid environments is still a difficult task. There is a lack of high level programming paradigms and support tools that may guide the application developer and allow reusability of state-of-the-art solutions. Specifically, the main goal of the work presented in this thesis is to contribute to the simplification of the development cycle of applications for Grid environments by bringing structure and flexibility to three stages of that cycle through a commonmodel. The stages are: the design phase, the execution phase, and the reconfiguration phase. The common model is based on the manipulation of patterns through pattern operators, and the division of both patterns and operators into two categories, namely structural and behavioural. Moreover, both structural and behavioural patterns are first class entities at each of the aforesaid stages. At the design phase, patterns can be manipulated like other first class entities such as components. This allows a more structured way to build applications by reusing and composing state-of-the-art patterns. At the execution phase, patterns are units of execution control: it is possible, for example, to start or stop and to resume the execution of a pattern as a single entity. At the reconfiguration phase, patterns can also be manipulated as single entities with the additional advantage that it is possible to perform a structural reconfiguration while keeping some of the behavioural constraints, and vice-versa. For example, it is possible to replace a behavioural pattern, which was applied to some structural pattern, with another behavioural pattern. In this thesis, besides the proposal of the methodology for distributed application development, as sketched above, a definition of a relevant set of pattern operators was made. The methodology and the expressivity of the pattern operators were assessed through the development of several representative distributed applications. To support this validation, a prototype was designed and implemented, encompassing some relevant patterns and a significant part of the patterns operators defined. This prototype was based in the Triana environment; Triana supports the development and deployment of distributed applications in the Grid through a dataflow-based programming model. Additionally, this thesis also presents the analysis of a mapping of some operators for execution control onto the Distributed Resource Management Application API (DRMAA). This assessment confirmed the suitability of the proposed model, as well as the generality and flexibility of the defined pattern operators

Calculating an upper bound on the finishing time of a group of threads executing on a GPU: a preliminary case study

Graphics processor units (GPUs) today can be used for computations that go beyond graphics and such use can attain a performance that is orders of magnitude greater than a normal processor. The software executing on a graphics processor is composed of a set of (often thousands of) threads which operate on different parts of the data and thereby jointly compute a result which is delivered to another thread executing on the main processor. Hence the response time of a thread executing on the main processor is dependent on the finishing time of the execution of threads executing on the GPU. Therefore, we present a simple method for calculating an upper bound on the finishing time of threads executing on a GPU, in particular NVIDIA Fermi. Developing such a method is nontrivial because threads executing on a GPU share hardware resources at very fine granularity.

Online robust optimization framework for QoS guarantees in distributed soft real-time systems

In distributed soft real-time systems, maximizing the aggregate quality-of-service (QoS) is a typical system-wide goal, and addressing the problem through distributed optimization is challenging. Subtasks are subject to unpredictable failures in many practical environments, and this makes the problem much harder. In this paper, we present a robust optimization framework for maximizing the aggregate QoS in the presence of random failures. We introduce the notion of K-failure to bound the effect of random failures on schedulability. Using this notion we define the concept of K-robustness that quantifies the degree of robustness on QoS guarantee in a probabilistic sense. The parameter K helps to tradeoff achievable QoS versus robustness. The proposed robust framework produces optimal solutions through distributed computations on the basis of Lagrangian duality, and we present some implementation techniques. Our simulation results show that the proposed framework can probabilistically guarantee sub-optimal QoS which remains feasible even in the presence of random failures.

The daily and hourly energy consumption and load forecasting using artifitial neural network method: a case study using a set of 93 households in Portugal

It is important to understand and forecast a typical or a particularly household daily consumption in order to design and size suitable renewable energy systems and energy storage. In this research for Short Term Load Forecasting (STLF) it has been used Artificial Neural Networks (ANN) and, despite the consumption unpredictability, it has been shown the possibility to forecast the electricity consumption of a household with certainty. The ANNs are recognized to be a potential methodology for modeling hourly and daily energy consumption and load forecasting. Input variables such as apartment area, numbers of occupants, electrical appliance consumption and Boolean inputs as hourly meter system were considered. Furthermore, the investigation carried out aims to define an ANN architecture and a training algorithm in order to achieve a robust model to be used in forecasting energy consumption in a typical household. It was observed that a feed-forward ANN and the Levenberg-Marquardt algorithm provided a good performance. For this research it was used a database with consumption records, logged in 93 real households, in Lisbon, Portugal, between February 2000 and July 2001, including both weekdays and weekend. The results show that the ANN approach provides a reliable model for forecasting household electric energy consumption and load profile. © 2014 The Author.

The extended unsymmetric frontal solution for multiple-point constraints

The purpose of this paper is to discuss the linear solution of equality constrained problems by using the Frontal solution method without explicit assembling. Design/methodology/approach - Re-written frontal solution method with a priori pivot and front sequence. OpenMP parallelization, nearly linear (in elimination and substitution) up to 40 threads. Constraints enforced at the local assembling stage. Findings - When compared with both standard sparse solvers and classical frontal implementations, memory requirements and code size are significantly reduced. Research limitations/implications - Large, non-linear problems with constraints typically make use of the Newton method with Lagrange multipliers. In the context of the solution of problems with large number of constraints, the matrix transformation methods (MTM) are often more cost-effective. The paper presents a complete solution, with topological ordering, for this problem. Practical implications - A complete software package in Fortran 2003 is described. Examples of clique-based problems are shown with large systems solved in core. Social implications - More realistic non-linear problems can be solved with this Frontal code at the core of the Newton method. Originality/value - Use of topological ordering of constraints. A-priori pivot and front sequences. No need for symbolic assembling. Constraints treated at the core of the Frontal solver. Use of OpenMP in the main Frontal loop, now quantified. Availability of Software.

On the robustness of the slotine-Li and the FPT/SVD-based adaptive controllers

A comparative study concerning the robustness of a novel, Fixed Point Transformations/Singular Value Decomposition (FPT/SVD)-based adaptive controller and the Slotine-Li (S&L) approach is given by numerical simulations using a three degree of freedom paradigm of typical Classical Mechanical systems, the cart + double pendulum. The effects of the imprecision of the available dynamical model, presence of dynamic friction at the axles of the drives, and the existence of external disturbance forces unknown and not modeled by the controller are considered. While the Slotine-Li approach tries to identify the parameters of the formally precise, available analytical model of the controlled system with the implicit assumption that the generalized forces are precisely known, the novel one makes do with a very rough, affine form and a formally more precise approximate model of that system, and uses temporal observations of its desired vs. realized responses. Furthermore, it does not assume the lack of unknown perturbations caused either by internal friction and/or external disturbances. Its another advantage is that it needs the execution of the SVD as a relatively time-consuming operation on a grid of a rough system-model only one time, before the commencement of the control cycle within which it works only with simple computations. The simulation examples exemplify the superiority of the FPT/SVD-based control that otherwise has the deficiency that it can get out of the region of its convergence. Therefore its design and use needs preliminary simulation investigations. However, the simulations also exemplify that its convergence can be guaranteed for various practical purposes.

Data analytics in the cloud with flexible mapreduced workflows

Data analytic applications are characterized by large data sets that are subject to a series of processing phases. Some of these phases are executed sequentially but others can be executed concurrently or in parallel on clusters, grids or clouds. The MapReduce programming model has been applied to process large data sets in cluster and cloud environments. For developing an application using MapReduce there is a need to install/configure/access specific frameworks such as Apache Hadoop or Elastic MapReduce in Amazon Cloud. It would be desirable to provide more flexibility in adjusting such configurations according to the application characteristics. Furthermore the composition of the multiple phases of a data analytic application requires the specification of all the phases and their orchestration. The original MapReduce model and environment lacks flexible support for such configuration and composition. Recognizing that scientific workflows have been successfully applied to modeling complex applications, this paper describes our experiments on implementing MapReduce as subworkflows in the AWARD framework (Autonomic Workflow Activities Reconfigurable and Dynamic). A text mining data analytic application is modeled as a complex workflow with multiple phases, where individual workflow nodes support MapReduce computations. As in typical MapReduce environments, the end user only needs to define the application algorithms for input data processing and for the map and reduce functions. In the paper we present experimental results when using the AWARD framework to execute MapReduce workflows deployed over multiple Amazon EC2 (Elastic Compute Cloud) instances.

A fractional calculus perspective in the evolutionary design of combinational circuits

This paper analyses the performance of a genetic algorithm (GA) in the synthesis of digital circuits using two novel approaches. The first concept consists in improving the static fitness function by including a discontinuity evaluation. The measure of variability in the error of the Boolean table has similarities with the function continuity issue in classical calculus. The second concept extends the static fitness by introducing a fractional-order dynamical evaluation.

Riesz potential operators and inverses via fractional centred derivatives

International Journal of Mathematics and Mathematical Sciences, Vol.2006

A pattern language for parallelizing irregular algorithms

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para obtenção do grau de Mestre em Engenharia Informática

Fuzzy modelling of prescribed burning effects on soil physical properties

This paper presents the preliminary work of an approach where Fuzzy Boolean Nets (FBN) are being used to extract qualitative knowledge regarding the effect of prescribed fire burning on soil chemical physical properties. FBN were chosen due to the scarcity on available quantitative data.

The computational role of short-term plasticity and the balance of excitation and inhibition in neural microcircuits: experimental and theoretical analysis

The computations performed by the brain ultimately rely on the functional connectivity between neurons embedded in complex networks. It is well known that the neuronal connections, the synapses, are plastic, i.e. the contribution of each presynaptic neuron to the firing of a postsynaptic neuron can be independently adjusted. The modulation of effective synaptic strength can occur on time scales that range from tens or hundreds of milliseconds, to tens of minutes or hours, to days, and may involve pre- and/or post-synaptic modifications. The collection of these mechanisms is generally believed to underlie learning and memory and, hence, it is fundamental to understand their consequences in the behavior of neurons.(...)