Multi-resource scheduling for FPGA systems

In modern cloud data centers, reconfigurable devices (FPGAs) are used as an alternative to Graphics Processing Units to accelerate data-intensive computations (e.g., machine learning, image and signal processing). Currently, FPGAs are configured to execute fixed workloads, repeatedly over long periods of time. This conflicts with the needs, proper to cloud computing, to flexibly allocate different workloads and to offer the use of physical devices to multiple users. This raises the need for novel, efficient FPGA scheduling algorithms that can decide execution orders close to the optimum in a short time. In this context, we propose a novel scheduling heuristic where groups of tasks that execute together are interposed by hardware reconfigurations. Our contribution is based on gathering tasks around a high-latency task that hides the latency of tasks, within the same group, that run in parallel and have shorter latencies. We evaluated our solution on a benchmark of 37500 random workloads, synthesized from realistic designs (i.e., topology, resource occupancy). For this testbench, on average, our heuristic produces optimum makespan solutions in 47.4% of the cases. It produces acceptable solutions for moderately constrained systems (i.e., the deadline falls within 10% of the optimum makespan) in 90.1% of the cases.     

EMA-PRBDS: efficient multi-attribute packet rank based data scheduling in wireless sensor networks for real-time monitoring systems

ABSTRACT

In recent years, Wireless Sensor Networks (WSNs) are widely placed in real-time monitoring systems like environmental, structural, patient monitoring, etc. The major criterion for WSN includes energy efficiency and network lifetime. Scheduling is used as a large number of data packets focus on the same queue at the same time. Only limited data scheduling schemes have been implemented in WSN to enhance the performance. The existing First Come First Serve (FCFS) and Dynamic Multilevel Priority (DMP) have some technical challenges like delay, packet drop and high energy consumption due to starvation and deadlock. In this paper, we proposed a new algorithm to optimise and rank the incoming data packets based on multi-attributes decision-making methodology named as Packet Rank Based Data Scheduling (PRBDS), the algorithm selects packet priority, deadline, and size as the metrics to rank the incoming data packets. A simulation result shows when compared with existing scheduling, PRBDS not only provides less energy consumption, also significantly reduces the packet drop and increases the lifetime. Thus, the proposed algorithm is most suitable for real-time monitoring system since it combines data ranking method with scheduling algorithm to create accurate and reliable results to evaluate the incoming data packets.     

Adaptive sampling rate control for networked systems based on statistical characteristics of packet disordering

This paper investigates an adaptive sampling rate control scheme for networked control systems (NCSs) subject to packet disordering. The main objectives of the proposed scheme are (a) to avoid heavy packet disordering existing in communication networks and (b) to stabilize NCSs with packet disordering, transmission delay and packet loss. First, a novel sampling rate control algorithm based on statistical characteristics of disordering entropy is proposed; secondly, an augmented closed-loop NCS that consists of a plant, a sampler and a state-feedback controller is transformed into an uncertain and stochastic system, which facilitates the controller design. Then, a sufficient condition for stochastic stability in terms of Linear Matrix Inequalities (LMIs) is given. Moreover, an adaptive tracking controller is designed such that the sampling period tracks a desired sampling period, which represents a significant contribution. Finally, experimental results are given to illustrate the effectiveness and advantages of the proposed scheme.     

GPRS技术在燃气监测系统改造的应用

针对数传电台在燃气监测系统运行中存在的问题,论述了用GPRS网络替代数传电台的改造。     

交通荷载下路面振动响应信号的时频特征分析

本文在简单分析了Fourier变换、短时Fourier变换和小波变换的基础上,以实测的交通荷载路面振动信号为模拟信号,采用小波包对模拟信号进行时频特征分析,并结合Fourier变换,给出了小波包各分解重构信号随时间的衰减规律及其在各子频带的频谱图,因此,结合Fourier变换对信号进行小波包分析是一种有效的信号处理途径.研究结果对今后交通荷载作用下路面路基动力响应计算模型的初始边界条件的确定提供依据,为路面路基动力响应的计算奠定基础.     

Reconfigurable control as actuatorfault-tolerant control design for poweroscillation damping

This paper presents design of an self contained actuators unit in wide area damping control of power system in stabilizing system response for both nominal system condition and during actuator faults. First it is presented that use of multiple actuators in wide area control aid in improving damping in power system. A wide area damping controller feeding multiple actuators to satisfy multiple objectives in wide area damping control of power system is designed. Minimization of infinity norm of closed loop transfer function of power system with wide area controller in feedback path & closed loop poles placement techniques are used in controller synthesis. Second a reconfigurable control on the lines of fault hiding principle is added to the controller design to maintain system damping to pre-fault level in case of actuator faults. A reconfiguration component(RC) is activated on occurrence of actuator fault thereby reconfiguring system dynamics and redistributing wide area control signal among remaining active actuators. RC together with remaining active actuators and under same wide area damping controller maintains system damping to pre-fault level thereby preserving system dynamic response. In the reconfigurable control design presented here no new actuators outside the unit of actuators designed for wide area damping control is required. This makes for an self contained actuators unit in wide area damping control of power system both for nominal system condition and for system affected by actuator faults. A two area power system model is considered here for demonstrating effectiveness of designed robust damping controller with multiple outputs feeding multiple actuators in wide area control and illustrating the idea of self contained actuators unit for maintaining system damping in case of actuator faults.     

The effect of system configuration and ramp-up time on manufacturing system acquisition under uncertain demand

The objective of this paper is to investigate the optimal allocation of capacity investments at the tactical decision-making level by incorporating the configuration characteristics of selected system alternatives comprising Dedicated Manufacturing Systems (DMS) and Reconfigurable Manufacturing Systems (RMS). Particularly, sequencing of stages in a series or a parallel configuration impacts the responsiveness in addressing capacity change requirements. We analyze what type of configuration is more suitable for a manufacturer in terms of service level and cost. We propose a mixed integer programming model by incorporating various ramp-up time patterns, which define system scalability lead time. By solving the MIP model to optimality, we aim to see how capacity is allocated to RMS and DMS based on system cost, system responsiveness, and reconfiguration speed. A discrete event simulation model is used to validate the MIP results under uncertain demand scenarios.     

A dynamic multi-objective approach for the reconfigurable multi-facility layout problem

The multi-facility layout problem involves the physical organization of departments inside several facilities, to allow flexible and efficient operations. This work studies the facility layout problem in a new perspective, considering a group of facilities, and two different concerns: the location of departments within a group of facilities, and the location of departments inside each facility itself. The problem is formulated as a Quadratic Programming Problem with multiple objectives and unequal areas, allowing layout reconfigurations in each planning period. The objectives of the model are: the minimization of costs (material handling inside facilities and between facilities, and re-layout); the maximization of adjacency between departments; and the minimization of the “unsuitability” of department positions and locations. This unsuitability measure is a new objective proposed in this work, to combine the characteristics of existing locations with the requirements of departments. The model was tested with data from the literature as well as with a problem inspired in a first tier supplier in the automotive industry. Preliminary results show that this work can be viewed as an innovative and promising integrated approach for tackling real, complex facility layout problems.     

Adaptive controller placement for wireless sensor–actuator networks with erasure channels

Wireless sensor–actuator networks offer flexibility for control design. One novel element which may arise in networks with multiple nodes is that the role of some nodes does not need to be fixed. In particular, there is no need to pre-allocate which nodes assume controller functions and which ones merely relay data. We present a flexible architecture for networked control using multiple nodes connected in series over analog erasure channels without acknowledgments. The control architecture proposed adapts to changes in network conditions, by allowing the role played by individual nodes to depend upon transmission outcomes. We adopt stochastic models for transmission outcomes and characterize the distribution of controller location and the covariance of system states. Simulation results illustrate that the proposed architecture has the potential to give better performance than limiting control calculations to be carried out at a fixed node.     

Scalable reconfigurable equipment design principles

Scalability is one of six key characteristics found in reconfigurable manufacturing systems. Scalable systems satisfy changing capacity requirements efficiently through system reconfiguration, and in the flexible manufacturing literature this capability is called expansion flexibility. The development of modular scalable machine tools is a necessary precursor to achieving scalable systems. Unfortunately, there is little work describing the design of scalable machines. This paper establishes the need for scalable machines and a basis for evaluating and describing them. Applicable metrics are defined, and an architecture for scalable machines is presented. Two examples illustrate the scalable architecture. Finally, a design parameter based on a mathematical approach is presented to determine the optimal number of modules to be included on a modular scalable machine. This as a design parameter is important because it limits machine size and the number of module interfaces included in the base machine structure.