This paper presents a static scheduler to carry out the best assignment of a Directed Acyclic Graph (DAG) representing an application program. Some characteristics of the DAG, a decision model, and the evaluation para...
详细信息
ISBN:
(纸本)3540440496
This paper presents a static scheduler to carry out the best assignment of a Directed Acyclic Graph (DAG) representing an application program. Some characteristics of the DAG, a decision model, and the evaluation parameters for choosing the best solution provided by the selected scheduling algorithms are defined. The selection of the scheduling algorithms is based on five decision levels. At each level, a subset of scheduling algorithms is selected. When the scheduler was tested with a series of DAGs having different characteristics, the scheduler's decision was right 100% of the time in those cases in which the number of available processors is known.
We consider a queueing network in which there axe constraints on which queues may be served simultaneously. Such networks, called "switched networks" (3] can be used to model input-queued switches, wireless ...
详细信息
ISBN:
(纸本)9781424429257
We consider a queueing network in which there axe constraints on which queues may be served simultaneously. Such networks, called "switched networks" (3] can be used to model input-queued switches, wireless networks, or bandwidth sharing in the Internet. The scheduling algorithm for such a network specifies which queues to serve at any point in time. The performance of scheduling algorithm is determined by the induced net queue-size. The question of designing optimal scheduling algorithm with this performance metric has been of great recent interest (e.g. [1, 3, 4]). An important step in this quest is that of finding fundamental limitations of scheduling algorithms in terms of the induced queue-size. In this paper, we present a novel technique to characterize lower bound on average queue-size induced by any algorithm. Through an example, we establish the tightness of this technique for a class of problems.
The elastic resource provision, non-interfering resource sharing and flexible customized configuration provided by the Cloud infrastructure has shed light on efficient execution of many scientific applications. Due to...
详细信息
ISBN:
(纸本)9781479950690
The elastic resource provision, non-interfering resource sharing and flexible customized configuration provided by the Cloud infrastructure has shed light on efficient execution of many scientific applications. Due to the increasing deployment of data centers and computer servers around the globe escalated by the higher electricity price, the energy cost on running the computing, communication and cooling together with the amount of CO2 emissions have skyrocketed. In order to maintain sustainable Cloud computing facing with ever-increasing problem complexity and big data size in the next decades, we design and develop energy-aware scientific workflow scheduling algorithm to minimize energy consumption and CO2 emission while still satisfying certain Quality of Service (QoS) such as response time specified in Service Level Agreement (SLA). We also apply Dynamic Voltage and Frequency Scaling (DVFS) and DNS scheme to further reduce energy consumption within acceptable performance bounds. Our multiple-step resource provision and allocation algorithm achieves the response time requirement in the step of forwarding task scheduling and minimizes the VM overhead for reduced energy consumption and higher resource utilization rate in the backward task scheduling step. The effectiveness of our algorithm is evaluated under various performance metrics and experimental scenarios using software adapted from open source CloudSim simulator.
Computer systems are becoming increasingly limited by memory performance. A dynamic SDRAM access scheduler (DSAS) according to modem SDRAM technology and memory access scheduling algorithms is proposed in this paper. ...
详细信息
ISBN:
(纸本)9781424411313
Computer systems are becoming increasingly limited by memory performance. A dynamic SDRAM access scheduler (DSAS) according to modem SDRAM technology and memory access scheduling algorithms is proposed in this paper. Based on SDR-SDRAM technology and new version AMBA AXI bus, DSAS dynamically schedules the accesses to SDRAM and reduces precharge time. A modularized configurable automatic verification platform is established to verify basic functions of DSAS. We analyze the results and conclude that the memory controller using DSAS is capable of predicting future operations, thus greatly hide the precharge time, which takes much time in SDRAM operations;AXI bus throughput rate has improved by 19%similar to 52% during frequent SDRAM accesses.
The cost efficiency of model inference is critical to real-world machine learning (ML) applications, especially for delay-sensitive tasks and resource-limited devices. A typical dilemma is: in order to provide complex...
详细信息
ISBN:
(纸本)9781577358763
The cost efficiency of model inference is critical to real-world machine learning (ML) applications, especially for delay-sensitive tasks and resource-limited devices. A typical dilemma is: in order to provide complex intelligent services (e.g. smart city), we need inference results of multiple ML models, but the cost budget (e.g. GPU memory) is not enough to run all of them. In this work, we study underlying relationships among black-box ML models and propose a novel learning task: model linking. Model linking aims to bridge the knowledge of different black-box models by learning mappings (dubbed model links) between their output spaces. Based on model links, we developed a scheduling algorithm, named MLink. Through collaborative multi-model inference enabled by model links, MLink can improve the accuracy of obtained inference results under the cost budget. We evaluated MLink on a multi-modal dataset with seven different ML models and two real-world video analytics systems with six ML models and 3,264 hours of video. Experimental results show that our proposed model links can be effectively built among various black-box models. Under the budget of GPU memory, MLink can save 66.7% inference computations while preserving 94% inference accuracy, which outperforms multi-task learning, deep reinforcement learning-based scheduler and frame filtering baselines.
Coverage redundancy problem is one of the significant problems in wireless sensor networks. To reduce the energy consumption that arises when the high number of sensors is active, various coverage control protocols (s...
详细信息
ISBN:
(纸本)9781424464043
Coverage redundancy problem is one of the significant problems in wireless sensor networks. To reduce the energy consumption that arises when the high number of sensors is active, various coverage control protocols (sleep scheduling algorithms) have been proposed. In these protocols, a subset of nodes necessary to maintain sufficient sensing coverage are kept active while the others are put into a sleep mode to reduce the energy consumption. In this paper, we study the coverage redundancy problem in a sensor network where the locations of nodes and the distances between nodes are neither known nor could be easily calculated. We define a neighbor graph as the graph formed by the neighbors of a node and analyze the effect of different levels of connectivity in neighbor graphs on the coverage redundancy of sensor nodes. Moreover, we apply our results to a lightweight deployment-aware scheduling algorithm and demonstrate the improvement in the performance of the algorithm.
A sensor grid is an integration of two technologies: wireless sensor networks and the grid. The sensors deployed in a WSN monitor a phenomenon of interest. The information gained from the WSN is processed in the grid ...
详细信息
ISBN:
(纸本)9781467348836;9781467348812
A sensor grid is an integration of two technologies: wireless sensor networks and the grid. The sensors deployed in a WSN monitor a phenomenon of interest. The information gained from the WSN is processed in the grid and is used by the users of applications. Multipurpose WSNs have become very popular where the deployed WSNs support more than one application. The research in this extended abstract focuses on WSNs supporting multiple applications. In this work, we focus on allocation, which is a process of determining the sensor nodes that will be selected for executing the requests corresponding to an application. scheduling, which determines the order in which the application requests submitted to the WSN are executed is performed to improve the mean response time to the users of the applications. Our previous works propose various scheduling algorithms for WSNs hosting multiple applications. In this research, various static and dynamic allocation algorithms are proposed with an attempt to balance the energy consumption amongst the sensor nodes and hence improve the network lifetime of the WSN. Network lifetime is the time when the energy of any sensor node in the WSN falls below a predefined threshold. The proposed algorithms use varying degree of information about the energy consumption at the major energy consuming components of the sensor nodes: the CPU component and the radio component. Simulation experiments are performed to evaluate the performance of the proposed algorithms. This extended abstract presents the preliminary results obtained from the experimentation done so far. The simulation experiments demonstrate that by performing dynamic allocation and by using information about the total energy consumption at the sensor nodes, the lifetime of the WSN can be significantly improved.
In this paper we consider deterministic differential equation models for the varying number of flows in a network. These arise naturally as limits of stochastic models under joint scaling of flow arrival rates and net...
详细信息
ISBN:
(纸本)1424403499
In this paper we consider deterministic differential equation models for the varying number of flows in a network. These arise naturally as limits of stochastic models under joint scaling of flow arrival rates and network capacities. We compare these dynamics under (i) coordinated multipath routing and (ii) parallel, uncoordinated routing. We show that for identical traffic demands, parallel uncoordinated routing can be unstable while balanced multipath routing is stable. In other words, coordination can strictly increase the schedulable region, that is the set of demand vectors for which the system is stable. We also show that, even when uncoordinated multipath routing stabilises the system, coordination can bring further benefits, as it naturally minimises network costs at equilibrium.
We present a sequencing problem given on JIT (Just In Time) manufacturing environments, with the objective of minimizing the variation of manufacturing rates (ORV: Output Rate Variation). Specifically, we propose an e...
详细信息
ISBN:
(纸本)9783642406423;9783642406430
We present a sequencing problem given on JIT (Just In Time) manufacturing environments, with the objective of minimizing the variation of manufacturing rates (ORV: Output Rate Variation). Specifically, we propose an extension based on requiring to the sequences the preservation of the production mix throughout the products manufacturing. To solve the ORV and the extended problem, we propose algorithms based on BDP (Bounded Dynamic Programming) and we perform two computational experiments based on instances from the literature.
This paper presents a novel FPGA-based switch design that achieves high algorithmic performance and an efficient FPGA implementation. Crossbar switches based on virtual output queues (VOQs) and variations have been ra...
详细信息
ISBN:
(纸本)9781665420105
This paper presents a novel FPGA-based switch design that achieves high algorithmic performance and an efficient FPGA implementation. Crossbar switches based on virtual output queues (VOQs) and variations have been rather popular for implementing switches on FPGAs, with applications to network-on-chip (NoC) routers and network switches. The efficiency of VOQs is well-documented on ASICs, though we show that their disadvantages can outweigh their advantages on FPGAs. Our proposed design uses an output-queued switch internally for simplifying scheduling, and a queue balancing technique to avoid queue fragmentation and reduce the need for memory-sharing VOQs. Our implementation approaches the scheduling performance of the state-of-the-art, while requiring considerably fewer FPGA resources.
暂无评论