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.

Efficient computation of min and max sensor values in multihop networks

Consider a wireless sensor network (WSN) where a broadcast from a sensor node does not reach all sensor nodes in the network; such networks are often called multihop networks. Sensor nodes take individual sensor readings, however, in many cases, it is relevant to compute aggregated quantities of these readings. In fact, the minimum and maximum of all sensor readings at an instant are often interesting because they indicate abnormal behavior, for example if the maximum temperature is very high then it may be that a fire has broken out. In this context, we propose an algorithm for computing the min or max of sensor readings in a multihop network. This algorithm has the particularly interesting property of having a time complexity that does not depend on the number of sensor nodes; only the network diameter and the range of the value domain of sensor readings matter.

Analyzing TDMA with slot skipping

Distributed real-time systems, such as factory automation systems, require that computer nodes communicate with a known and low bound on the communication delay. This can be achieved with traditional time division multiple access (TDMA). But improved flexibility and simpler upgrades are possible through the use of TDMA with slot-skipping (TDMA/SS), meaning that a slot is skipped whenever it is not used and consequently the slot after the skipped slot starts earlier. We propose a schedulability analysis for TDMA/SS. We assume knowledge of all message streams in the system, and that each node schedules messages in its output queue according to deadline monotonic. Firstly, we present a non-exact (but fast) analysis and then, at the cost of computation time, we also present an algorithm that computes exact queuing times.

Exact analysis of TDMA with slot skipping

Consider a communication medium shared among a set of computer nodes; these computer nodes issue messages that are requested to be transmitted and they must finish their transmission before their respective deadlines. TDMA/SS is a protocol that solves this problem; it is a specific type of Time Division Multiple Access (TDMA) where a computer node is allowed to skip its time slot and then this time slot can be used by another computer node. We present an algorithm that computes exact queuing times for TDMA/SS in conjunction with Rate-Monotonic (RM) or Earliest- Deadline-First (EDF).

The capacity exchange protocol

This paper proposes a new strategy to integrate shared resources and precedence constraints among real-time tasks, assuming no precise information on critical sections and computation times is available. The concept of bandwidth inheritance is combined with a capacity sharing and stealing mechanism to efficiently exchange bandwidth among tasks to minimise the degree of deviation from the ideal system’s behaviour caused by inter-application blocking. The proposed Capacity Exchange Protocol (CXP) is simpler than other proposed solutions for sharing resources in open real-time systems since it does not attempt to return the inherited capacity in the same exact amount to blocked servers. This loss of optimality is worth the reduced complexity as the protocol’s behaviour nevertheless tends to be fair and outperforms the previous solutions in highly dynamic scenarios as demonstrated by extensive simulations. A formal analysis of CXP is presented and the conditions under which it is possible to guarantee hard real-time tasks are discussed.

Exact admission-control for integrated aperiodic and periodic tasks

Admission controllers are used to prevent overload in systems with dynamically arriving tasks. Typically, these admission controllers are based on suÆcient (but not necessary) capacity bounds in order to maintain a low computational complexity. In this paper we present how exact admission-control for aperiodic tasks can be eÆciently obtained. Our rst result is an admission controller for purely aperiodic task sets where the test has the same runtime complexity as utilization-based tests. Our second result is an extension of the previous controller for a baseload of periodic tasks. The runtime complexity of this test is lower than for any known exact admission-controller. In addition to presenting our main algorithm and evaluating its performance, we also discuss some general issues concerning admission controllers and their implementation.

Music and evolutionary computation

This paper presents a brief history of the western music: from its genesis to serialism and the Darmstadt school. Also some mathematical aspects of music are then presented and confronted with music as a form of art. The question is, are these two distinct aspects compatible? Can computers be of real help in automatic composition? The more appealing algorithmic approach is evolutionary computation as it offers creativity potential. Therefore, the Evolutionary Algorithms are then introduced and some results of GAs and GPs application to music generation are analysed.

Uma metaheurística para a programação de projectos com multi-modos e recursos limitados

Este artigo apresenta uma nova abordagem (MM-GAV-FBI), aplicável ao problema da programação de projectos com restrições de recursos e vários modos de execução por actividade, problema conhecido na literatura anglo-saxónica por MRCPSP. Cada projecto tem um conjunto de actividades com precedências tecnológicas definidas e um conjunto de recursos limitados, sendo que cada actividade pode ter mais do que um modo de realização. A programação dos projectos é realizada com recurso a um esquema de geração de planos (do inglês Schedule Generation Scheme - SGS) integrado com uma metaheurística. A metaheurística é baseada no paradigma dos algoritmos genéticos. As prioridades das actividades são obtidas a partir de um algoritmo genético. A representação cromossómica utilizada baseia-se em chaves aleatórias. O SGS gera planos não-atrasados. Após a obtenção de uma solução é aplicada uma melhoria local. O objectivo da abordagem é encontrar o melhor plano (planning), ou seja, o plano que tenha a menor duração temporal possível, satisfazendo as precedências das actividades e as restrições de recursos. A abordagem proposta é testada num conjunto de problemas retirados da literatura da especialidade e os resultados computacionais são comparados com outras abordagens. Os resultados computacionais validam o bom desempenho da abordagem, não apenas em termos de qualidade da solução, mas também em termos de tempo útil.

On the numerical computation of the Mittag-Leffler function

Recently simple limiting functions establishing upper and lower bounds on the Mittag-Leffler function were found. This paper follows those expressions to design an efficient algorithm for the approximate calculation of expressions usual in fractional-order control systems. The numerical experiments demonstrate the superior efficiency of the proposed method.

On the numerical computation of the Mittag-Leffler function

Recently simple limiting functions establishing upper and lower bounds on the Mittag-Leffler function were found. This paper follows those expressions to design an efficient algorithm for the approximate calculation of expressions usual in fractional-order control systems. The numerical experiments demonstrate the superior efficiency of the proposed method.

An Exact Schedulability Test for Global FP Using State Space Pruning

Presented at 23rd International Conference on Real-Time Networks and Systems (RTNS 2015). 4 to 6, Nov, 2015, Main Track. Lille, France.

A congestion management and transmission price simulator for competitive electricity markets

In a liberalized electricity market, the Transmission System Operator (TSO) plays a crucial role in power system operation. Among many other tasks, TSO detects congestion situations and allocates the payments of electricity transmission. This paper presents a software tool for congestion management and transmission price determination in electricity markets. The congestion management is based on a reformulated Optimal Power Flow (OPF), whose main goal is to obtain a feasible solution for the re-dispatch minimizing the changes in the dispatch proposed by the market operator. The transmission price computation considers the physical impact caused by the market agents in the transmission network. The final tariff includes existing system costs and also costs due to the initial congestion situation and losses costs. The paper includes a case study for the IEEE 30 bus power system.