Search

The current study investigated the moderating roles of age and trust in the relation of procedural justice with turnover. It was expected that the relation between procedural justice and turnover was weaker for older workers and those with high prior trust in their leader. Older workers are better at regulating their emotions, and focus more on positive aspects of their relationships with others, and therefore react less intensely to unfair treatment. Moreover, people with high trust are more likely to attribute unfair treatment to circumstances instead of deliberate intention than people with low trust. Finally, we expected a three-way interaction between age, trust, and procedural justice in relation to turnover, where older workers with high trust would have less strong reactions than younger workers and older workers with low trust. Results from a three-wave longitudinal survey among 1,597 Dutch employees indeed revealed significant interactions between trust and procedural justice in relation to turnover. Furthermore, the three-way interaction was significant, with negative relations for younger workers, but a non-significant relation was found for older workers with low trust. Contrary to expectations, negative relations were found between procedural justice and turnover for older workers with high trust. © 2010 The Authors. Journal compilation © 2010 International Association of Applied Psychology.

Multimedia applications, executed by embedded multiprocessor systems, can in some cases be represented as task graphs, with the tasks containing nested loop programs. The nested loop programs communicate via arrays and can be executed on different processors. Typically an array can be communicated via a circular buffer with a capacity smaller than the array. For such buffers, the communicating nested loop programs have to synchronize and a sufficient buffer capacity needs to be computed. In a circular buffer we use a write and a read window to support rereading, out-of-order reading or writing, and skipping of locations. A cyclo static dataflow model is derived from the application and used to compute buffer capacities that guarantee deadlock free execution. Our case-study applies circular buffers in a Digital Audio Broadcasting channel decoder application, where the frequency deinterleaver reads according to a non-affine pseudo-random function. For this application, buffer capacities are calculated that guarantee deadlock free execution.

In traditional wireless sensor network (WSN) applications, energy efficiency is considered to be the most important concern whereas utilizing the use of bandwidth and maximizing the throughput are of secondary importance. However, recent applications, such as structural health monitoring, require high amounts of data to be collected at a faster rate. We present a multi-channel MAC protocol, MC-LMAC, designed with the objective of maximizing the throughput of WSN by coordinating transmissions over multiple channels. MC-LMAC takes the advantage of interference and contention free parallel transmissions on different channels. It is based on scheduled access which eases the coordination of nodes dynamically switching their interfaces between channels and makes the protocol to operate effectively free of collisions during peak traffic. Time is organized into timeslots and each node shall be assigned control over a timeslot to transmit on a particular channel. We analyze the performance of MC-LMAC with extensive simulations. MC-LMAC exhibits significant bandwidth utilization and high throughput while ensuring an energy-efficient operation. Moreover, MC-LMAC outperforms the contention-based multi-channel MMSN protocol, cluster based channel assignment and single-channel CSMA in terms of data delivery ratio and throughput for high data rate, moderate-size networks of 100 nodes.

In modern multiprocessor systems, processors can be stalled by inter-task communication when reading from a remote buffer. This paper presents a solution for the inter-task communication, that has a minimal impact on the performance of the system, hides the inter-task communication latency without requiring additional hardware. The solution applies to jobs, represented as task graphs, where the tasks are nested loop programs. Buffers are allocated in scratch-pad memories of the consuming tasks to provide low latency read access. For the nested loop programs, minimal buffer sizes can be determined to cover all possible communication patterns. The added computational complexity is low, as the solution adds only a few operations to the nested loop programs.

Our Multi Processor System on Chip (MPSoC) template provides processing tiles that are connected via a network on chip. A processing tile contains a processing unit and a Scratch Pad Memory (SPM). This paper presents the Omphale tool that performs the first step in mapping a job, represented by a task graph, to such an MPSoC, given the SPM sizes as constraints. Furthermore a memory tile is introduced. The result of Omphale is a Cyclo Static DataFlow (CSDF) model and a task graph where tasks communicate via sliding windows that are located in circular buffers. The CSDF model is used to determine the size of the buffers and the communication pattern of the data. A buffer must fit in the SPM of the processing unit that is reading from it, such that low latency access is realized with a minimized number of stall cycles. If a task and its buffer exceed the size of the SPM, the task is examined for additional parallelism or the circular buffer is partly located in a memory tile. This results in an extended task graph that satisfies the SPM size constraints.

We have investigated the radio interference behavior of a multi-channel WSN system with a typical radio platform. We have analyzed the results in terms of throughput and throughput per-channel spacing. The experimental results show that we can adjust the channel spacing with respect to the level of tolerated interference according to the application requirements. This gives us the possibility of simultaneous usage of as many as channels possible according to the permitted interference.

This paper describes an algorithm to determine the performance of real-time systems with tasks using stochastic processing times. Such an algorithm can be used for guaranteeing Quality of Service of periodic tasks with soft real-time constraints. We use a discrete distribution model of processing times instead of worst case times like in hard real-time systems. Such a model gives a more realistic view on the actual requirements of the system. The presented algorithm works for all deterministic scheduling systems, which makes it more general than existing 6algorithms and allows us to compare performance between these systems. To demonstrate our method, we make a comparison between the performance of the well known scheduling algorithms Earliest Deadline First and Rate Monotonic. We show that the complexity of our method can compete with other algorithms that work for a wide range of schedulers.

With the growing popularity of Wireless Sensor Networks (WSNs), the demand increases to perform time critical operations within such networks. A WSN is composed of sensor nodes, which contain a radio, ports for multiple sensors and a microcontroller. These sensor nodes have to provide a wide range of functionality as long as possible, while they use their scarce energy from a battery.

A key step in the design of cyclo-static real-time systems is the determination of buffer capacities. In our multiprocessor system, we apply back-pressure, which means that tasks wait for space in output buffers. Consequently buffer capacities affect the throughput. This requires the derivation of buffer capacities that both result in a satisfaction of the throughput constraint, and also satisfy the constraints on the maximum buffer capacities. Existing exact solutions suffer from the computational complexity that is associated with the required conversion from a cyclo-static dataflow graph to a single-rate dataflow graph. In this paper we present an algorithm, with linear computational complexity, that does not require this conversion and that strives to obtain close to minimal buffer capacities. The algorithm is applied to an MP3 playback application that is mapped on our multi-processor system.

Currently, most wireless sensor network applications assume the presence of single-channel Medium Access Control (MAC) protocols. When sensor nodes are densely deployed, single-channel MAC protocols may be inadequate due to the higher demand for the limited bandwidth. To overcome this drawback, we propose multiple channel support for improving the performance. Our method allows the nodes to utilize new frequency channels which results in the significant increase on the number of nodes that are granted access to the wireless medium. The method requires only one half-duplex transceiver per node, which is capable of sending and receiving over distinguished frequency channels. Simulation results show that, method successfully utilizes multiple channels and increases the performance proportional to the number of available frequencies for an example single-channel MAC protocol, LMAC.

A key step in the design of multi-rate real-time systems is the determination of buffer capacities. In our multi-processor system, we apply back-pressure as caused by bounded buffers in order to control jitter. This requires the derivation of buffer capacities that both satisfy the temporal constraints as well as constraints on the buffer capacity. Existing exact solutions suffer from the computational complexity associated with the required conversion from a multi-rate dataflow graph to a single-rate dataflow graph. In this paper we present an algorithm, with linear computational complexity, that does not require this conversion and that determines close to minimal buffer capacities. The algorithm is applied to an MP3 play-back application that is mapped on our network based multi-processor system.

We propose an approach for providing Quality-of-Service guarantees in a virtual channel Network-on-Chip. The approach is based on virtual channel reservation - routes with guaranteed lower throughput bound are reserved over the virtual channels. The performance of such an approach is limited by a number of factors: number of virtual channel in the network, number of requested routes, traffic locality etc. We test the performance of the proposed approach for variety of traffic conditions. We investigate the influence of three system parameters - routing algorithm, network topology and communication locality - on the performance limits of the approach. The results are derived by simulations using a streaming traffic model. The results show the approach is feasible for a network of size 10-by-10 with four virtual channels per physical channel. The traffic locality has strong influence on the performance limits of the approach and can also help in reducing the communication energy cost by 50% to 70%. The type of the routing algorithm does not practically influence the performance limits.

This paper examines the possibilities of providing throughput guarantees in a network-on-chip by appropriate traffic routing. A source routing function is used to find routes with specified throughput for the data streams in a streaming multiprocessor system-on-chip. The influence of the routing algorithm, network topology and communication locality on the routing performance are studied. The results show that our method for providing throughput guarantees to streaming traffic is feasible. The communication locality has the strongest influence on the routing performance while the routing algorithm has weakest influence. Therefore, the mapping algorithm is of greater importance for the system performance than the routing algorithm and it is profitable to use a more complex mapping algorithm that preserves the communication locality together with a simple routing algorithm.

We describe a self-organizing, clustering protocol for bandwidth resource management in Wireless Sensor Networks. The proposed protocol allows the sensor nodes to communicate by a time-slotted, scheduled MAC algorithm. When the nodes are densely deployed, i.e., the connectivity is very high, the MAC algorithm may not provide access for all of the nodes due to the limited number of time-slots, consequently the network capacity degrades. To overcome this drawback, we extend the time-slotted MAC algorithm by clustering the nodes into direrent frequency domains while they can use the same time domain. The idea is basically to multiplex the time domain with the frequency domain. As a result, the number of nodes that are granted access to the wireless medium is increased by the number of frequency channels available. By using simulations, we evaluate the performance of the protocol. The results reveal that frequency multiplexing has the erect of increasing the capacity up to 100%.

This work is an initiative attempt toward component-based software engineering in ubiquitous computing systems. Software components cooperate in a distributed manner to meet a demand, and adapt their software bindings during run-time depending on the context information. There are two main research topics investigated in this study. The first topic is how to build an architecture, consisting of software components, that supports adaptation and self-configuration. We explain why that component is needed, what the requirements are and how it will be designed. Besides component-based design, we build our architecture on top of Publish/Subscribe (P/S) model. We like to reuse the TinyOS components and as a second topic, we investigate the problems that will be experienced when converting these components to our P/S oriented environment. Our experiences during this research pointed out that; buffer exchange and flexible interface name representation are not suitable for ubiquitous systems.

In this paper we study the timing aspects of the operation of stream-processing applications that run on a multiprocessor architecture. Dependencies are derived for the processing and communication times of the processors in such a system. Three cases of real-time constrained operation and four cases of communication organization are considered and compared. Examples of application are given for the derived results.

DCOS is a Data Centric lightweight Operating System for embedded devices. Despite limited energy and hardware resources, it supports a data driven architecture with provisions for dynamic loadable Modules. It combines these with Real-Time provisions based on Earliest Deadline First with a simple but smart resource handling mechanism. We will give an overview of the capabilities of DCOS and we will describe the basics of the main mechanisms.

The real-time token-based RTnet network protocol has been implemented on a standard Ethernet network to investigate the possibility to use cheap components with strict resource limitations while preserving Quality of Service guarantees. It will be shown that the proposed implementation is feasible on a small network. For larger networks a different approach is necessary, using delegation by means of proxies. A delegation proposal will be discussed. For small networks it is possible to use a PIC microcontroller in combination with a standard Ethernet controller to run the RTnet network protocol. As more systems are added to the network the performance of this combination becomes insufficient. When this happens it is necessary for the microcontroller to delegate some tasks to a more powerful master and to organize a low-level communication protocol between master and slave.

This paper compares the implementation results of two architectures for virtual channel router. Since the router is used for building an on-chip network, its small size is critical. Together with the total design area we provide information about the distribution of this area between the main router blocks and thus give insight about the cost of each block. The comparison shows that one of the architectures results in smaller implementation area and overcomes some performance problems presented by the other architecture.

Promote-IT is an efficient heuristic scheduler that provides QoS guarantees for accessing data from tertiary storage. It can deal with a wide variety of requests and jukebox hardware. It provides short response and confirmation times, and makes good use of the jukebox resources. It separates the scheduling and dispatching functionality and effectively uses this separation to dispatch tasks earlier than scheduled, provided that the resource constraints are respected and no task misses its deadline. To prove the efficiency of Promote-IT we implemented alternative schedulers based on different scheduling models and scheduling paradigms. The evaluation shows that Promote-IT performs better than the other heuristic schedulers. Additionally, Promote-IT provides response-times near the optimum in cases where the optimal scheduler can be computed.

This paper describes a medium access protocol for real-time communication in wireless networks. Medium access is controlled by a scheduler, which utilizes a pre-emptive earliest deadline first (PEDF) scheduling algorithm. The scheduler prevents collisions in the network, where normally only collisions are avoided. PEDF is the scheduler of choice, because it has excellent properties with respect to bandwidth utilization, dynamic behaviour and feasibily analysis. The scheduler can be deployed in managed networks, where it resides in the base station, as well as in peer-to-peer or ad hoc networks, were it is distributed over the stations. The protocol is simulated and an implementation based on IEEE 802.11b is realized.

This paper describes several existing data link layer protocols that provide real-time capabilities on wired networks, focusing on token-ring and Carrier Sense Multiple Access based networks. Existing modifications to provide better real-time capabilities and performance are also described. Finally the pros and cons regarding the At-Home Anywhere project are discussed.

This paper focuses on communication in personal area networks. A personal area networks (PAN) is characterized as an informal collection, or community, of connected small, lightweight, and resource-lean devices, or gadgets. Two basic concepts are visible in the development of PANs, the distributed and the centralized concept. The paper introduces a real-time communication protocol that is suitable for both concepts. The communication protocol can deal with several types of traffic: real-time or nonreal- time, bursty or isochronous, high or low bitrate. The protocol is undemanding in terms of resources, so even simple devices can participate in the network. The network is simulated and a prototype is realized.

RTN is a network access protocol that uses a token scheduling mechanism to enable real-time multimedia streaming on Ethernet-like networks. Unlike other token based protocols, RTNs token mechanism is based on pre-emptive earliest deadline first (PEDF) scheduling. PEDF has interesting characteristics, such as one hundred percent bandwidth utilization and a uncomplicated feasibility analysis. RTN is simulated and implemented on Ethernet and IEEE802.11. The latter implementation is not straightforward, because some for RTN important details are not defined in the standard. As a consequence the standard had to be evaluated before adaptation and implementation of RTN could start. This paper briefly describes the RTN protocol. The focus, however, is on benchmarking the IEEE802.11 standard.

We present a distributed clock synchronization algorithm. It performs clock synchronization on an Ethernet-based real-time token local area network, without the use of an external clock source. It is used to enable the token schedulers in each node to agree upon a common time. Its intended use is in resource-lean systems, where heavyweight protocols like NTP cannot be used. We present a short overview of the working of the protocol, as well as experimental results.

EDFI is a lightweight real-time scheduling protocol that combines EDF with deadline inheritance over shared resources. We will show that EDFI is flexible during a tasks admission control, efficient with scheduling and dispatching, and straightforward in feasibility analysis. The application programmer only needs to specify a tasks timing constraints (deadline, period, runtime) and resource needs, after which EDFI can execute admission control, scheduling, dispatching and resource synchronisation automatically. EDFI avoids gratuitous task switching and its programming overhead as well as runtime overhead is very low, which makes it ideal for lightweight and featherweight kernels. We will illustrate the elegance of the underlying theory and we will shortly discuss the implementation of EDFI in three different operating systems.

When shared resources are involved, scheduling in current hard real-time operating systems, too often has its timely behaviour guaranteed at the cost of a rather complicated administration. We will show that we can improve this considerably by using methods based on so-called Real-Time Transactions (RTTs). A RTT is a task that has guaranteed the use of all needed resources after it has started, without ever having to wait for the resource release; an RTT is only started if these resources are free. RTTs allow for a complete separation of a real-time application and its involved system support. Scheduling, (shared) resource synchronisation and admission control is executed automatically by the underlying system, while an application programmer only needs to specify timing constraints (deadline, period, runtime) and resource needs. We will discuss the implementation of RTTs within Plan 9 as used at Bell-labs, and we will illustrate the straightforward and elegant use of our transaction scheduling theory.

This paper presents the jukebox early quantum scheduler (JEQS). JEQS is a periodic jukebox scheduler for a Video-on-Demand system. JEQS uses the jukebox robots in a cyclic way and the time is divided in constant units called quanta. A quantum is the maximum time needed to unload and load all the drives. An RSM is loaded in a drive for a fixed period of time, corresponding to the time needed to switch the media on the other drives. During this time the drive can read data from it. JEQS is based on the scheduling theory on early quantum tasks (EQT). An early quantum task executes its first instance in the next quantum after its arrival and the rest of the instances are scheduled in a normal periodic way with the release time immediately after the first execution. Although JEQS is an efficient periodic scheduler, that can guarantee the execution of most tasks in the next cycle after the requests arrive, we show that using JEQS results in much longer response times than using a-periodic schedulers. Furthermore, we show that the bad performance of JEQS is intrinsic to any periodic jukebox scheduler. The only advantage of using a periodic scheduler is that the scheduling algorithms are less complex. However, the simplicity of the algorithms clearly does not outweigh the unacceptably long response times.

Scheduling jukebox resources is important to build efficient and flexible hierarchical storage systems. JukeTools is a toolbox that helps in the complex tasks of implementing and evaluating jukebox schedulers. It allows the fast development of jukebox schedulers. The schedulers can be tested in numerous environments, real and simulated. JukeTools helps the developer to easily detect errors in the schedules. Analyzer tools create detailed reports on the behavior and performance of any of the scheduler, and provide comparisons between different schedulers. This paper describes the functionality offered by JukeTools, with special emphasis on how the toolbox can be used to develop jukebox schedulers.

An Early Quantum Task (EQT) is a Quantum EDF task that has shrunk its first period into one quantum time slot. Its purpose is to be executed as soon as possible, without causing deadline overflow of other tasks. We will derive the conditions under which an EQT can be admitted and can have an immediate start. The advantage of scheduling EQTs is shown by its use in a buffered multi-media server. The EQT is associated with a multimedia stream and it will use its first invocation to fill the buffer, such that a client can start receiving data immediately.

This paper presents an analytical hardware model for tertiary-storage jukeboxes. A jukebox can store large amounts of data in a cost-effective way, which makes it eminently suitable for applications that handle large amounts of continuous-media files, large databases and backups. However, to use a jukebox effectively it is important to schedule the jukebox resources. In order to build accurate schedules we need good predictions of the jukebox performance, which are provided by the hardware model. We also use this model to simulate jukeboxes.

This paper shows the current state of our research into a home network which provides both real-time and non-real-time capabilities for one coherent, distributed architecture. It is based on a new type of real-time token protocol that uses scheduling to achieve optimal token-routing in the network. Depending on the scheduling algorithm, bandwidth utilisations of 100% are possible. Token management, to prevent token-loss or multiple tokens, is essential to support a dynamic, plug-and-play configuration. Our network will support inexpensive, small appliances as well as more expensive, large appliances. Small appliances, like sensors, would contain low-cost, embedded processors with limited computing power, which can handle lightweight network protocols. All other operations can be delegated to other appliances that have sufficient resources. This provides a basis for transparency, as it separates controlling and controlled object.

ClockWork shows that we can improve the flexibility and efficiency of real-time kernels. We do this by proposing methods for scheduling based on so-called Real-Time Transactions. ClockWork uses Real-Time Transactions which allow scheduling decisions to be taken by the system. A programmer does not need to be aware of synchronisation due to the sharing of resources and may have the illusion of a run-to-completion semantics even under pre-emptive scheduling protocols. The ClockWork tool presented here analyses the schedulability of a set of RT Transactions for a variety of protocols and visualises the result in a graphical form.

The design for a new type of non-volatile mass storage memory is discussed. This new design, based on scanning probe techniques, combines the low volume and power consumption of the FlashRAM, with the high capacity of the hard disk. The small form factor of the device makes it an excellent candidate for mass storage in handheld embedded systems. Its hierarchical architecture allows us to make a trade-off between data-rate, access time and power consumption. The power consumption scales linearly with the desired data-rate, and is expected to be lower than what can be achieved with competing technologies.

The design for a new type of non-volatile mass storage memory is discussed. This new design, based on scanning probe techniques, combines the low volume and power consumption of the FlashRAM, with the high capacity of the hard disk. The small form factor of the device makes it an excellent candidate for mass storage in handheld embedded systems. Its hierarchical architecture allows us to make a trade-off between data-rate, access time and power consumption. The power consumption scales linearly with the desired data-rate, and is expected to be lower than what can be achieved with competing technologies.

We present a jukebox scheduler for real-time data. The scheduler is part of a hierarchical real-time file system to be used over a network. A jukebox is a large tertiary storage device whose removable media (e.g. cd-rom, dvd-rom) are loaded and unloaded from one or more drives by a robot. The problem with tertiary storage is that media exchange times are high and the number of drives is limited. This makes scheduling tertiary storage complicated. The storage media switching time in a jukebox is in the order of tens of seconds. Therefore multiplexing between two files stored in different media is many orders of magnitude slower than doing the same in secondary storage. The goal of the scheduler is to schedule the use of the jukebox devices (arm and drives) in such a way that the system can guarantee the deadlines while minimizing the response time. The problem is similar to that of scheduling multiple processors with the additional difficulty of having to deal with the high switching times and the use of a shared resource (the arm). Finding an optimal schedule is an NP-hard problem. We provide a near-optimal polynomial solution by using heuristics to prune the tree of solutions. The scheduling time is in average less than 100 ms. The incoming requests are scheduled on-line.

Current Hard Real-Time (HRT) kernels have their timely behaviour guaranteed on the cost of a rather restrictive use of the available resources. This makes current HRT scheduling techniques inadequate for use in a multimedia environment where we can make a considerable profit by a better and more flexible use of the resources. We will show that we can improve the flexibility and efficiency of multimedia kernels. Therefore we introduce Real Time Transactions (RTT) with Deadline Inheritance policies for a small class of scheduling algorithms and we will evaluate these algorithms for use in a multimedia environment