Our results demonstrate that our novel application development environment provides both ease-of-use and high performance for large-scale, I/O-intensive scientific applications.
ISBN:
(纸本)9781581132700
Our results demonstrate that our novel application development environment provides both ease-of-use and high performance for large-scale, I/O-intensive scientific applications.
To achieve reliability, reusability and cost reduction, a significant trend of building large, complex real-time systems is to integrate separated application modules. An essential requirement of integrated real-time ...
详细信息
To achieve reliability, reusability and cost reduction, a significant trend of building large, complex real-time systems is to integrate separated application modules. An essential requirement of integrated real-time systems is to guarantee strong partitioning among applications. In this paper, we propose a microkernel, called the SPIRIT (Strongly Partitioned Integrated real-time sysTem) /spl mu/Kernel, for strongly partitioned real-time systems. The SPIRIT-/spl mu/Kernel has been designed and implemented based on a two-level hierarchical scheduling methodology such that the real-time constraints of each application can be guaranteed. It provides a minimal set of kernel functions, such as address management, interrupt/exception dispatching, inter-application communication and application scheduling. To demonstrate the feasibility of the SPIRIT-/spl mu/Kernel, we have ported two different application-level real-time operating systems (RTOSs)-WindRiver's VxWorks and Cygnus's eCos-on the top of the microkernel. The SPIRIT-/spl mu/Kernel architecture is practical and appealing due to its low overheads of kernel services and the support for dependable integration of real-time applications.
Motion control and arcade games require a high bandwidth and periodic transfers to meet the needs of more precise control of motors, and more players and more complex graphics in arcade games. There is a common-memory...
详细信息
As the component raises a principal factor in software engineering to ensure the productivity and reusability of the software, component extracting and implementation, component-based system construction, verification...
详细信息
We demonstrate how to utilize the path information available from ITS navigation systems for bandwidth reservation for hand-off and admission control of new connections in mobile cellular networks. By utilizing the pa...
详细信息
We demonstrate how to utilize the path information available from ITS navigation systems for bandwidth reservation for hand-off and admission control of new connections in mobile cellular networks. By utilizing the path information, we modify the recently proposed bandwidth reservation and admission control scheme with a design goal to keep the hand-off dropping probability below a pre-specified target. While the original scheme utilized a history-based mobility estimation, the modified scheme eases only part of it. Since both the original and modified schemes meet the design goal, the problem is to determine which of the two can admit more new connections for a given traffic condition. Using simulations, the modified scheme aided by the-navigation system is shown to outperform the original scheme even with much less computational requirement.
Motion control and arcade games require a high bandwidth and periodic transfers to meet the needs of more precise control of motors, and more players and more complex graphics in arcade games. There is a common-memory...
详细信息
Motion control and arcade games require a high bandwidth and periodic transfers to meet the needs of more precise control of motors, and more players and more complex graphics in arcade games. There is a common-memory architecture called "Cyclic Memory" because the data is refreshed periodically which is well known in the field of real-time systems. IEEE 1394 has been drawing considerable attention as the need for connecting audio/visual electronics with a unified fast interface increases. We discuss the feasibility of Cyclic Memory on the IEEE 1394 with its isochronous mode. By analyzing the requirements of two prototypical applications, we evaluate the performance and other capabilities of IEEE 1394 and the other networks that have the functionality of Cyclic Memory and discuss their applicability. We also discuss the network reliability based on IEEE 1394 when it is applied to an actual motion control system.
Unlike deterministic realtime communication in which excessive resources may be required for absolute performance guarantees, statistical realtime communication seeks to achieve both probabilistic performance guaran...
详细信息
Unlike deterministic realtime communication in which excessive resources may be required for absolute performance guarantees, statistical realtime communication seeks to achieve both probabilistic performance guarantees and efficient resource sharing. We propose a framework for statistical realtime communication in ATM networks that provides delay guaranteed transport of MPEG coded video traffic with a statistically guaranteed cell loss ratio. In order to provide delay guaranteed communication service, we employ a modified version of Traffic-Controlled Rate-Monotonic Priority Scheduling (TCRM). We multiplex a set of statistical realtime channels that share: (i) similar traffic characteristics into a common channel called a macro channel, and (ii) the resources of the macro channel. Individual statistical realtime channels are given timeliness and probabilistic cell loss guarantees. A macro channel is serviced by the modified TCRM which improves link utilization and makes channel management simpler. Using the analysis of an M/D/1/N queueing system, we propose a procedure for determining the transmission capacity of a macro channel needed to statistically guarantee a cell loss ratio bound. Simulation results have shown our framework to work well as compared to the other approaches. The overall cell loss ratios for multi hop statistical realtime channels are shown to be smaller than the pre-determined bounds.
We present a balanced and unbalanced binary-tree model to explore the delay performance of feedback synchronization algorithms for multicast ABR flow control. Using this model, we analyze the feedback-delay performanc...
详细信息
We present a balanced and unbalanced binary-tree model to explore the delay performance of feedback synchronization algorithms for multicast ABR flow control. Using this model, we analyze the feedback-delay performance scalability of the soft-synchronization protocol (SSP), which derives a single "consolidated" RM (resource management) cell at each multicast branch-point from feedback RM cells of different downstream branches that are not necessarily responses to the same forward RM cell. In contrast with the other existing schemes, SSP is shown to be able to effectively support synchronization of feedback RM cell and make the effective RM cell roundtrip time (RTT) virtually independent of the multicast tree's topology. Also derived is the optimal RM-cell update interval for SSP to minimize RM-cell RTTs for a given multicast tree.
Random early detection (RED) was proposed as an active gateway queue-management mechanism. This paper proposes to alter the RED design guideline that unconditionally allows transient congestion, and evaluates its impa...
详细信息
Random early detection (RED) was proposed as an active gateway queue-management mechanism. This paper proposes to alter the RED design guideline that unconditionally allows transient congestion, and evaluates its impacts. This unconditional allowance of transient congestion is shown to be harmful when the queue is near full, because it causes buffer overflow at a gateway. Buffer overflow at a gateway leads to the global synchronization and oscillation of traffic load on the network. To effectively prevent buffer overflow at a gateway, the RED framework is refined in such a way that the gateway can detect a transient congestion in a timely manner and take actions to quench it when the queue is near full. Based on our simulation results, two enhancements are made in estimating average queue size. Equipped with these enhancements, the refined RED can strike a good balance between the allowance of transient bursty traffic and the avoidance of buffer overflow. Using extensive simulations, the refined RED is comparatively evaluated against and is shown to be superior to the original RED.
暂无评论