Design of the shared memory system for multi-processor lisp machines and its implementation on the evlis machine

In multi-core systems, main memory is a major shared resource among processor cores. A task running on one core can be delayed by other tasks running simultaneously on other cores due to interference in the shared main memory system. Such memory interference delay can be large and highly variable, thereby posing a significant challenge for the design of predictable real-time systems. In this paper, we present techniques to reduce this interference and provide an upper bound on the worst-case interference on a multi-core platform that uses a commercial-off-the-shelf (COTS) DRAM system. We explicitly model the major resources in the DRAM system, including banks, buses, and the memory controller. By considering their timing characteristics, we analyze the worst-case memory interference delay imposed on a task by other tasks running in parallel. We find that memory interference can be significantly reduced by (i) partitioning DRAM banks, and (ii) co-locating memory-intensive tasks on the same processing core. Based on these observations, we develop a memory interference-aware task allocation algorithm for reducing memory interference. We evaluate our approach on a COTS-based multi-core platform running Linux/RK. Experimental results show that the predictions made by our approach are close to the measured worst-case interference under workloads with both high and low memory contention. In addition, our memory interference-aware task allocation algorithm provides a significant improvement in task schedulability over previous work, with as much as 96 % more tasksets being schedulable.

A solid fertilizer and seed application rate measuring system for a seed-fertilizer drill machine

Work done for the research project SATURNE is summarized. The technique developed in this project maximizes the redundancy level of tasks and tolerates hardware faults by majority voting in the context of a pool of interconnected processors (network of workstations, multiprocessor machines, etc.). The technique, called saturation, is presented and briefly compared with similar approaches. The distributed scheduling algorithm implementing this technique is described by using formal description techniques based on Estelle. Description analysis involves available simulation tools from which results were obtained and used as input parameters to the model defined for evaluation. >

This paper reports results of a study to determine and compare the effectiveness of various techniques aimed at reducing memory interference in multiprocessor multi memory systems. A simulation model of a multiprocessor, driven by address traces, is used for evaluation purposes. Two approaches to interference reduction are considered, being the reduction of overlapping memory requirements through various memory allocation methods, and the use of local memory to reduce the request rate to shared memory. Both private and cache memory are considered for this purpose.

Discrete element model for performance analysis of cutterhead excavation system of EPB machine

Despite attractiveness of multicore processors for embedded systems, the potential performance gains need to be studied in the context of real-time task scheduling and memory interference. This paper explores performance-aware schedulability of multicore systems by evaluating the performance when changing scheduling policies (as design parameters). The model-based framework we build enables analyzing the performance of multicore time-critical systems using processor-centric and memory-centric scheduling policies. The system architecture we consider consists of a set of cores with a local cache and sharing the cache level L2 and main memory (DRAM). The metrics we use to compare the performance achieved by different configurations of a system are: 1) utilization of the cores; and 2) the maximum delay per access request to shared cache and DRAM. Our framework, realized using Uppaal, can be viewed as an engineering tool to be used during design stages to identify the scheduling policies that provide better performance for a given system while maintaining system schedulability. As a proof of concept, we analyze and compare 2 different cases studies.

Silicon Photo-Multiplier (SiPM) detectors represent an attractive solution for the detection of low energy photons in several fields of both high energy physics and medical imaging. We present here an accurate electrical model for this kind of detectors, which can be conveniently used to perform reliable simulations at circuit level. A suitable extraction procedure for the parameters involved in the model is also described, based on both static and dynamic measurements. The proposed model allows to reproduce accurately the waveform of the signal generated by the SiPM when coupled to the front-end electronics, as shown by excellent fittings obtained between simulations and measurements taken on real devices. This is particularly useful in order to choose the most suitable front-end architecture for SiPM detectors, since the performance of the whole detection system, especially in terms of dynamic range and timing resolution, can be correctly predicted as a function of the detector parameters and of the main characteristics of the coupled electronics.

This paper introduced tilting accident phenomenon of the farm grasping grass machine in the practical application. Through to analyze overturn accident of instantaneous critical stress state at work, this paper put forward a basic idea that can control and prevent the overturn accident, designed the snatch and holding up safety automatic control system. It can detect the pressure changes of lifting hydraulic by installed pressure sensor in the lifting oil cylinder, and with the aid of weight-pressure mathematical model, cooperate with dynamic testing technology measures and corresponding measures of hardware, to realize the dynamic real-time monitoring of snatch and holding up weight. That the function of accumulated and recorded load, the early warning of overweight and accident prevention about falling down are realized by the system overall structure and software design.

In this paper, firstly, based on the application model of optimal scheduling of machining system for green manufacturing, two application models of the energy-saving scheduling model and energy-saving and noise-reducing scheduling model in a multi-model framework are combined, and the resource environment coefficient matrix of the two application models is established as well as the solution process is studied with the parallel machine problem. Then the system's architecture is constructed, and its basic operation flow, functional modules, etc., are designed and conceived. The application of the system is studied in conjunction with a gear machining workshop of a machine tool factory, and the machining system's energy and noise reduction performance is verified based on experiments. The results show that the energy consumption of the machining system is reduced by 0.514 kW-h by machining only the above six gear parts with a small difference in the maximum machining completion time and that the spindle speed has the most significant effect on the machine tool machining noise at a significance level α of 0.05. The analysis of this study verifies that the energy-saving and noise-reducing scheduling arrangement method can reduce the system machining energy consumption and noise, which is important for green manufacturing.

Reasonable ratio between the load inertia and servo motor inertia plays a decisive role for the dynamic performance and stability of the servo system, as well as the machining accuracy of the whole CNC machine. In order to improve the control performance and contour machining accuracy of the servo system of the CNC cycloidal gear form grinding machine, an optimization design method of the inertia matching for the CNC cycloidal gear form grinding machine servo system is proposed. The two-mass servo driving closed-loop PID control system is constructed, the influence of the different inertia ratios on the dynamic performance and contour errors of the servo system are deeply analyzed, and the inertia ratio is optimized to satisfied with the servo system performance requirements. Finally, the feasibility and practicability of the optimization design method of inertia matching are verified through the inertia ratio optimization grinding experiments of the cycloid gear in the CNC gear form grinding machine. This inertia matching optimization design method provides a valuable reference for the further design of CNC machine servo system.

Multiple-disk I/O systems (disk arrays) have been an attractive approach to meet high performance I/O demands in data intensive applications such as information retrieval systems. When we partition and distribute files across multiple disks to exploit the potential for I/O parallelism, a balanced I/O workload distribution becomes important for good performance. Naturally, the performance of a parallel information retrieval system using an inverted file structure is affected by the partitioning scheme of the inverted file. In this paper, we propose two different partitioning schemes for an inverted file system for a shared-everything multiprocessor machine with multiple disks. We study the performance of these schemes by simulation under a number of workloads where the term frequencies in the documents are varied, the term frequencies in the queries are varied, the number of disks are varied and the multiprogramming level is varied. >

Dynamic strain signals are important for many structural monitoring applications, but the high-speed interrogation of strain sensors based on fibre Bragg gratings (FBGs) remains a challenge. Arrayed waveguide grating (AWG) interrogation schemes have been proposed, and by using modelling and validation experiments several design considerations are investigated and their relationship to system performance indicators determined. The Bragg grating length has an impact on the ability of the grating to ‘observe’ the transient strain field, while the spectral widths of both the FBG and AWG influence the recovered strain resolution. The system performance was examined for both high frequency noise as well as the long-term drifts over an hour, with a strain resolution of 1.4 µε observed and drift of less than 3.1 µε h−1. The noise dependence on the relative overlap of the AWG and FBG spectra was found to be significant and the inclusion of a semiconductor optical amplifier to boost light intensity at the detectors was found to significantly improve performance with an improvement in the signal-to-noise ratio up to 200%.

An approach for networked three-layer water tank control system is advanced using a predictive compensation strategy to adapt the detrimental effect of networked QoS induced by communication networks．Firstly,a dynamic on-line measurement method is presented for typical networked QoS performance．Secondly,a compensation strategy is designed for the given networked QoS and system performance targets．Finally,a relevant software is developed and tested through campus-wide net．The experimental results illustrate the effectiveness of the proposed scheme．

This paper investigates the gains harvested through base station cooperation in the up-link for a multi-user (MU) Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) system, operating in a real indoor environment. The base stations perform joint detection using an iterative receiver that carries out multi-user detection and channel estimation via soft information from the single-user decoders. Performance evaluation is carried out using real channels from an indoor dynamic dual MIMO link measurement campaign. The measured scenario represent a real life situation where two users communicate with two base stations, each with two antennas, in an environment resembling a shopping mall or an airport terminal. System performance is evaluated in terms of both Bit-Error Rate (BER) vs. Signal-to-Interference Ratio (SIR) and Cumulative Distribution Functions (CDF) for the instantaneous BER. Also, the impact of using soft information in the channel estimation is analyzed.

This study presents a schematic approach to model, simulate and analyze the overall performance of an FMS, comprising a carousel layout surrounded by manufacturing and assembly cells capable to deliver a variety of multiple products. All manufacturing and assembly cells are consisting of multiple multifunctional production and assembly machines, served by material loading and unloading robots, connected through a material transporting conveyor. Inter-cellular and intra-cellular routing flexibility are employed in the system through production resources operational and breakdown modes. Colored Petri net tools are used to develop a compact, editable and deadlock-free flexible manufacturing system model reveals important information about dynamic behavior and system performance measures e.g.; throughput, mean cycle time, and work-in-process (WIP) with reference to material handling and supply system, production resources reliability and process execution input variables. Variation in input factors namely mean machining and assembly time, material loading and unloading time, number of operations between failure, repair time, parts inter-arrival time, conveyor speed and buffer capacity have shown a substantial effect on the system performance. The results indicate successful applicability of colored Petri net tools for modelling, simulation, and evaluation of carousel configured multiple products flexible manufacturing system. Further, the optimal solution represents a deep relationship between the different production modes and responses to maximize the performance of a carousel configured multiple products FMS.