Parameter-sweep has been widely adopted in large numbers of scientific applications. Parameter-sweep features need to be incorporated into grid workflows so as to increase the scale and scope of such applications. New...
详细信息
Parameter-sweep has been widely adopted in large numbers of scientific applications. Parameter-sweep features need to be incorporated into grid workflows so as to increase the scale and scope of such applications. New scheduling mechanisms and algorithms are required to provide optimized policy for resource allocation and task arrangement in such a case. This paper addresses scheduling sequential parameter-sweep tasks in a fine-grained manner. The optimization is produced by pipelining the subtasks and dispatching each of them onto well-selected resources. Two types of scheduling algorithms are discussed and customized to adapt the characteristics of parameter-sweep, as well as their effectiveness has been compared under multifarious scenarios.
We present an algorithm for scheduling distributed data intensive bag-of-task applications on data grids that have costs associated with requesting, transferring and processing datasets. We evaluate the algorithm on a...
详细信息
ISBN:
(纸本)9780780390379
We present an algorithm for scheduling distributed data intensive bag-of-task applications on data grids that have costs associated with requesting, transferring and processing datasets. We evaluate the algorithm on a data grid testbed and present the results.
Formal specification languages are often perceived as difficult to use by practitioners, and are therefore rarely-used in industrial software development practices. Numerous researchers have developed specification pa...
详细信息
Formal specification languages are often perceived as difficult to use by practitioners, and are therefore rarely-used in industrial software development practices. Numerous researchers have developed specification pattern systems to facilitate the construction of formal specifications of system properties. Feedback indicates that these patterns are considered helpful, but many practitioners prefer capturing properties using informal notations, such as natural language, instead of formal specification languages. This paper describes a project that addresses this technology gap. First, we introduce a stepwise process for deriving and instantiating system properties in terms of their natural language representations. The key components of this process are structured natural language grammars and specification pattern systems. Second, we describe SPIDER, a prototype implementation of a tool suite supporting this specification process. We illustrate the use of our approach with a description of a stepwise construction process of property specifications of a real-world automotive embedded system using Spider.
Reprogramming of sensor networks is an important and challenging problem as it is often necessary to reprogram the sensors in place. In this paper, we propose a multihop reprogramming service designed for Mica-2/XSM m...
详细信息
Reprogramming of sensor networks is an important and challenging problem as it is often necessary to reprogram the sensors in place. In this paper, we propose a multihop reprogramming service designed for Mica-2/XSM motes. One of the problems in reprogramming is the issue of message collision. To reduce the problem of collision and hidden terminal problem, we propose a sender selection algorithm that attempts to guarantee that in a neighborhood there is at most one source transmitting the program at a time. Further, our sender selection is greedy in that it tries to select the sender that is expected to have the most impact. We also use pipelining to enable fast data propagation. MNP is energy efficient because it reduces the active radio time of a sensor node by putting the node into "sleep" state when its neighbors are transmitting a segment that is not of interest. Finally, we argue that it is possible to tune our service according to the remaining battery level of a sensor, i.e., it can be tuned so that the probability that a sensor is given the responsibility of transmitting the code is proportional to its remaining battery life
Feedback control real-time scheduling (FCS) aims at satisfying performance specifications of real-time systems based on adaptive resource management. Existing FCS algorithms often rely on the existence of continuous c...
详细信息
Feedback control real-time scheduling (FCS) aims at satisfying performance specifications of real-time systems based on adaptive resource management. Existing FCS algorithms often rely on the existence of continuous control variables in real-time systems. A number of real-time systems, however, support only a finite set of discrete configurations that limit the adaptation mechanisms. This paper presents hybrid supervisory utilization control (HySUCON) for scheduling such real-time systems. HySUCON enforces processor utilization bounds by managing the switchings between the discrete configurations. Our approach is based on a best-first-search algorithm that is invoked only if reconfiguration is necessary. Theoretical analysis and simulations demonstrate that the approach leads to robust utilization bounds for varying execution times. Experimental results demonstrate the algorithm performance for a representative application scenario.
Over the last few years, grid technologies have progressed towards a service-oriented paradigm that enables a new way of service provisioning based on utility computing models. Users consume these services based on th...
详细信息
Over the last few years, grid technologies have progressed towards a service-oriented paradigm that enables a new way of service provisioning based on utility computing models. Users consume these services based on their QoS (quality of service) requirements. In such "pay-per-use" grids, workflow execution cost must be considered during scheduling based on users' QoS constraints. In this paper, we propose a cost-based workflow scheduling algorithm that minimizes execution cost while meeting the deadline for delivering results. It can also adapt to the delays of service executions by rescheduling unexecuted tasks. We also attempt to optimally solve the task scheduling problem in branches with several sequential tasks by modeling the branch as a Markov decision process and using the value iteration method
Security critical systems must perform at the required security level, make effective use of available resources, and meet end-users expectations. Balancing these needs, and at the same time fulfilling budget and time...
详细信息
Grid and peer-to-peer (P2P) network technologies enable aggregation of distributed resources for solving large-scale and computationally-intensive applications. These technologies are well-suited for bag-of-tasks (BoT...
详细信息
Grid and peer-to-peer (P2P) network technologies enable aggregation of distributed resources for solving large-scale and computationally-intensive applications. These technologies are well-suited for bag-of-tasks (BoT) applications, because each application consists of many parallel and independent tasks. With multiple users competing for the same resources, the key challenge is to finish a user application within a specified deadline. In this paper, we propose a time optimization algorithm that schedules a user application on auction-based resource allocation systems. These allocation systems, which are based on proportional share, allow users to bid higher in order to gain more resource shares. Therefore, this algorithm adjusts a user bid periodically on these systems in order to finish the application on time.
As a new unsupervised learning technique, manifold learning has captured the attention of many researchers in the field of machine learning and cognitive sciences. The major algorithms include Isometric mapping (ISOMA...
详细信息
As a new unsupervised learning technique, manifold learning has captured the attention of many researchers in the field of machine learning and cognitive sciences. The major algorithms include Isometric mapping (ISOMAP) and Locally Linear Embedding (LLE). The approaches can be used for discovering the intrinsic dimensions of nonlinear high-dimensional data effectively and aim researchers to analyze the data better. How to quantitatively analyze the relationship between the intrinsic dimensions and the observation space, however, has fewer reports. And thus further works in manifold learning may have suffered some difficulties. The paper focuses on two kinds of manifold learning algorithms (ISOMAP, LLE), and discusses magnification factors and principal spread directions from the observation space to the intrinsic low-dimensional space. Also the corresponding algorithm is proposed. Experiments show the effectiveness and advantages of the research.
Security critical systems must perform at the required security level, make effective use of available resources, and meet end-users expectations. Balancing these needs, and at the same time fulfilling budget and time...
详细信息
Security critical systems must perform at the required security level, make effective use of available resources, and meet end-users expectations. Balancing these needs, and at the same time fulfilling budget and time-to-market constraints, requires developers to design and evaluate alternative security treatment strategies. In this paper, the authors presented a development framework that utilizes Bayesian belief networks (BBN) and aspect-oriented modeling (AOM) for a cost-benefit trade-off analysis of treatment strategies. AOM allows developers to model pervasive security treatments separately from other system functionality. This eases the trade-off by making it possible to swap treatment strategies in and out when computing return on security investments (RoSI). The trade-off analysis is implemented using BBN, and RoSI is computed by estimating a set of variables describing properties of a treatment strategy. RoSI for each treatment strategy is then used as input to choice of design.
暂无评论