Heterogeneous computing systems composed of interconnected machines with varied computational capabilities often operate in environments where there may be sudden machine failures, higher than expected load, or inaccu...
详细信息
ISBN:
(纸本)0769523129
Heterogeneous computing systems composed of interconnected machines with varied computational capabilities often operate in environments where there may be sudden machine failures, higher than expected load, or inaccuracies in estimation of system parameters, Makespan (defined as the completion time for an entire set of tasks) is often the performance feature that is optimized in such systems. It is important that the makespan of a resource allocation (mapping) be robust against errors in task computation time estimates. The problem of optimally mapping tasks onto machines of a heterogeneous computing environment has been shown, in general, to be NP-complete. Therefore, heuristic techniques to find near optimal solutions to this mapping problem are required. The goal of this research is to find a static mapping of tasks so that the robustness of the desired system feature, makespan, is maximized against the errors in task execution time estimates. Seven heuristics to derive near-optimal solutions and an upper bound to this problem are presented and evaluated.
Workflow management systems (WfMSs) are software systems used to automate, coordinate and streamline business processes of organizations. Most existing research on WfMSs has been focused on workflows based on well-str...
详细信息
Workflow management systems (WfMSs) are software systems used to automate, coordinate and streamline business processes of organizations. Most existing research on WfMSs has been focused on workflows based on well-structured and well-behaved business processes. Although its importance has been recognized, only recently, modeling and handling of workflow exceptions has been tackled by the workflow community. In this paper, we propose a mobile agent based approach to handling exceptions in distributed workflow management systems. Mobile agents are dispatched to find out the status of running processes in the system to keep track and troubleshoot them when necessary. We describe the classification of workflow exceptions into hierarchical levels and the corresponding design of different types of exception-handling mobile agents and their cooperation. A prototype of the mobile agent-based workflow exception handling mechanism has been implemented using the IBM Aglet platform
Most of today's broadband networks, including the recently ratified IEEE 802.16 standard, employ reservation based multiple media access control. A problem pertinent to reservation MAC protocols is the division of...
详细信息
Most of today's broadband networks, including the recently ratified IEEE 802.16 standard, employ reservation based multiple media access control. A problem pertinent to reservation MAC protocols is the division of frame slots between the contention and data transmission processes. In most of the reservation MAC protocols no specific ratio is standardized, leaving proprietary solutions address the local network environment. As both processes are equally important for maintaining efficient delay and throughput performance, a solution must consider the timely varying traffic load. For example, heterogeneity and cooperation of networks promote access technologies that can sustain waves of increasing traffic load. In this paper, we start by instituting a framework for efficient allocation of frame resources to the contention and data transmission processes in light of the delay and throughout performance. We then propose a dynamic resource allocation controller based on a Markovian optimization model, where the optimization parameters are tuned according to specific preferential criteria of service providers. Our model achieves opportunistic performance improvements, on a per frame basis, over the best-case static allocation. Through simulation, we study the merits of our proposed optimized controller with respect to the framework. We show by illustrative examples and numerical results that the controller successfully fulfills the framework objectives.
Software defined radios have the potential of changing the fundamental usage model of wireless communications devices, but the capabilities of these transceivers are often limited by the speed of the underlying proces...
详细信息
Software defined radios have the potential of changing the fundamental usage model of wireless communications devices, but the capabilities of these transceivers are often limited by the speed of the underlying processors and FPGAs. This paper presents a testbed for the design of an impulse-based ultra wideband communication system. The transceiver is being developed using software/reconfigurable radio concepts and will be implemented using commercially available off-the-shelf components. The receiver uses eight 1 GHz ADCs to perform time interleaved sampling at an aggregate rate of 8 Gsamples/s. The high sampling rates present extraordinary demands on the down-conversion resources. The output of each ADC is in a different clock domain, with clocks offset in increments of 125 ps. Samples are captured by the high-speed ADC and processed using a Xilinx Virtex-II Pro (XC2VP70) FPGA. The testbed has two components: a non-real time part for data capture and signal acquisition, and a real-time part for data demodulation and signal processing. The non-real time component uses the internal block RAMs to store a set of samples and one of the PowerPC cores to process the data offline, to minimize logic resource usage. The real-time part uses distributed memory to store incoming data and processes it using hardwired multipliers and FPGA logic cells. The overall objective is to demonstrate a testbed that will allow researchers to evaluate different UWB modulation, multiple access, and coding schemes, and will support raw data rates of up to 100 MB/s.
Bandwidth adaptation algorithms (BAA) assume an integral role in radio resource management (RRM) frameworks for future wireless networks. Taking advantage of the nature of adaptive multimedia application, they empower...
详细信息
Bandwidth adaptation algorithms (BAA) assume an integral role in radio resource management (RRM) frameworks for future wireless networks. Taking advantage of the nature of adaptive multimedia application, they empower system providers with allocation versatility and economic increase. Due to the dynamic nature to be witnessed in future networks, BAA has been set to be trigger at every change in the system state. This burdens the different entities involved in RRM. In this work, the concept of stochastically triggered BAA (STBAA) is proposed. The tradeoff between admission ratios and users satisfaction is evaluated. It is found that considerable gains can still be achieved with the adaptation not fully engaged
A microwave spectrum analyzer capable of capturing multi-GHz spectra with sub-MHz resolution and unity probability of intercept based on optical spectral holebuming materials is proposed and initial demonstrations pre...
详细信息
As wireless communications advances, the complexity of waveform processing places an ever increasing demand on baseband processing resources. As this is a difficult problem for traditional processing solutions to addr...
详细信息
As wireless communications advances, the complexity of waveform processing places an ever increasing demand on baseband processing resources. As this is a difficult problem for traditional processing solutions to address, many have looked to reconfigurable computing as a solution. For 3G systems, whose intense processing requirements exacerbate these problems, we have previously proposed in (S. Srikanteswara et al, Soft Radio Impl. for 3G and Future High Data Rate Syst., Globecom 2001) the use of custom computing machines (CCMs), also known as adaptive computing machines (ACMs). This paper introduces a method for measuring processor performance which is independent of chip architecture. This method is then used to perform a comparative analysis on the suitability of Virginia Tech's Stallion CCM and Texas, Instrument's TMS320C6201 processor for use in 3G applications.
Unified Parallel C (UPC) is a programming model for shared-memory parallel computing on shared- and distributed-memory systems. The Berkeley UPC software, which operates on top of their Global Addressing Space Network...
详细信息
Unified Parallel C (UPC) is a programming model for shared-memory parallel computing on shared- and distributed-memory systems. The Berkeley UPC software, which operates on top of their Global Addressing Space Networking (GASNet) communication system, is a portable, high-performance implementation of UPC for large-scale clusters. The Scalable Coherent Interface (SCI), a torus-based system-area network (SAN), is known for its ability to provide very low latency transfers as well as its direct support for both shared-memory and message-passing communications. High-speed clusters constructed around SCI promise to he a potent platform for large-scale UPC applications. This work introduces the design of the core API for the new SCI conduit for GASNet and UPC, which is based on active messages (AM). Latency and bandwidth data were collected and are compared with raw SCI results and with other existing GASNet conduits. The outcome shows that the new GASNet SCI conduit is able to provide promising performance in support of UPC applications.
Aircraft networks are used to service mission-critical avionics systems as well as cabin systems such as in-flight entertainment. These networks require that the switches used offer line-rate switching as well as boun...
详细信息
Aircraft networks are used to service mission-critical avionics systems as well as cabin systems such as in-flight entertainment. These networks require that the switches used offer line-rate switching as well as bounded latency and jitter. Gigabit Ethernet offers an interesting replacement to traditional proprietary networks because of its high performance and low cost. We develop a framework for analyzing the abilities of Gigabit Ethernet switches to provide probabilistic guarantees for reliably low latency using quality of service (QoS) controls. From the perspective of control capabilities, management complexity, and implementation success we compare the strengths and weaknesses of three modern Ethernet switches. These switches exhibit a broad range of configuration options and are representative of the different levels of QoS implementation available today. We use the performance metrics of latency, jitter, and packet loss to characterize the success of a service policy. The results from a variety of network, load, and traffic scenarios are presented in terms of these metrics. Through the results, we are able to gain insight into the individual switch implementations of QoS.
暂无评论