Resource Reservation (resource + reservation)

Distribution by Scientific Domains

Selected Abstracts

Resource reservations with fuzzy requests

T. Röblitz
Abstract We present a scheme for reserving job resources with imprecise requests. Typical parameters such as the estimated runtime, the start time or the type or number of required CPUs need not be fixed at submission time but can be kept fuzzy in some aspects. Users may specify a list of preferences which guide the system in determining the best matching resources for the given job. Originally, the impetus for our work came from the need for efficient co-reservation mechanisms in the Grid where rigid constraints on multiple job components often make it difficult to find a feasible solution. Our method for handling fuzzy reservation requests gives the users more freedom to specify the requirements and it gives the Grid Reservation Service more flexibility to find optimal solutions. In the future, we will extend our methods to process co-reservations. We evaluated our algorithms with real workload traces from a large supercomputer site. The results indicate that our scheme greatly improves the flexibility of the solution process without having much affect on the overall workload of a site. From a user's perspective, only about 10% of the non-reservation jobs have a longer response time, and from a site administrator's view, the makespan of the original workload is extended by only 8% in the worst case. Copyright © 2006 John Wiley & Sons, Ltd. [source]

Threshold-based admission control for a multimedia Grid: analysis and performance evaluation

Yang Zhang
Abstract In a Grid-based services system facing a large number of requests with different services and profits significance, there is always a trade-off between the system profits and the Quality of Service (QoS). In such systems, admission control plays an important role: the system has to employ a proper strategy to make admission control decisions and reserve resources for the coming requests thus to achieve greater profits without violating the QoS of the requests already admitted. In this paper, we introduce three essential admission control strategies with threshold on resource reservation and a newly proposed strategy with layered threshold. Through comprehensive theoretical analyses and extensive simulations, we demonstrate that the strategy with layered threshold is more efficient and flexible than the existing strategies for Grid-based multimedia services systems. Copyright © 2006 John Wiley & Sons, Ltd. [source]

QoS in IntServ-based IP networks: the peak rate policing

Lorenzo Battaglia
In the last few years, IP has moved towards resource reservation, with the task to guarantee in the future Quality of Service (QoS). This has led to flow admission control algorithms based on the negotiation of standardised traffic parameters. QoS can be guaranteed in any network, a priori from the used technology, only if the used admission control algorithm wisely shares the network's resources among the users. Any admission control algorithm on its turn can do so, only if every user respects the negotiated traffic parameters. Since any user could, maliciously or not, send at a higher rate than negotiated, i.e. use a higher share of resources than the negotiated one, in every network in which admission control is performed, a policing algorithm is used. An ideal policer should guarantee to reject no packet of a well-behaved user and police contract violation as rigidly as possible. All this independently of the characteristics of the monitored stream and of the background traffic. This holds also for Integrated Services (IS) based IP networks. In these networks, every user negotiates a peak and an average rate. In this paper we present the solution to the peak rate policing issue. We adapt the Generic Cell Rate Algorithm (GCRA), well-known policer used in ATM networks, to police the peak rate of flows of packets with variable length. We intuitively call this modified GCRA Generic Packet Rate Algorithm (GPRA) and dimension its parameters so that independently of the characteristics of the policed flow and of the background traffic, no packets of a well-behaved user are rejected and that the flows of any misbehaving user are rigidly policed. Copyright © 2003 AEI. [source]

The study of handoff prediction schemes for resource reservation in mobile multimedia wireless networks

Li-Liann Lu
Abstract In this paper, the mobility-dependent predictive resource reservation (MDPRR) scheme is proposed to provide flexible usage of scarce resource in mobile multimedia wireless networks. An admission control scheme is also considered to further guarantee the QoS of real-time traffic. The area of a cell is divided into non-handoff, pre-handoff, and handoff zones so that bandwidth is reserved in the target/sub-target cell as mobile stations move into the pre-handoff zone and leave the serving base station. The amount of bandwidth to be reserved is dynamically adjusted according to the location, the instantaneous variation of velocity and direction of mobile stations. Two scenarios of the MDPRR scheme are compared by considering the velocity threshold in the calculation of the weight of direction. A number of designs are investigated to further enhance the performance of the proposed scheme. The results show that employing the velocity threshold in the MDPRR scheme can indeed reduce connection dropping probability, and make better usage of the reserved bandwidth. Copyright © 2004 John Wiley & Sons, Ltd. [source]

Using virtual topologies to manage inter-domain QoS in next-generation networks

Ricardo B. Freitas
Recently, several computer fields have turned to virtualization as a way to simplify complex problems. In this context, the Virtual Topology Service (VTS) was created to manage the advertisement and acquisition of virtual topologies (abstractions of the network status of a domain) and their use in inter-domain resource reservation to provide end-to-end quality of service (QoS). As an effort to create new network architectures which could attend current requirements like mobility and context-aware applications and support autonomous, heterogeneous and mobile domains next-generation networks (NGNs) emerged, with Ambient Networks (AN) as one of its instances. With an ever increasing multitude of online applications, end-to-end QoS has become increasingly important, especially for media and real-time uses. In this context, in order to better manage inter-domain QoS in these new networks, better coping with mobile nodes and domains, this work presents a new design and implementation of the VTS, adapted to the AN environment. The new VTS stores resource reservation information to enable the reuse of these reservations when re-establishing QoS after a node/domain movement. This implementation was tested on a real NGN prototype and showed considerable time saving when compared to QoS re-establishment without reusing the reservations. Copyright © 2009 John Wiley & Sons, Ltd. [source]

An adaptive resource reservation for vehicular mobile networks

I. Ben Hamida
This paper presents the time-based bandwidth reservation (TBR) algorithm, suitable for handoff management in cellular systems. TBR is based on real-time measurements of mobile stations (position, velocity and acceleration). The scheme consists in sending reservation requests to the neighboring cells based on an extrapolation of the user's motion. The originality of our approach lies in dynamically adjusting the amount of time for which bandwidth has to be allocated and reserved in a cell. In addition, we propose an optimal channel requests arrangement (CRA) algorithm in order to improve the performance of TBR in terms of resource utilization. Finally, we propose VTBR, an adapted and extended version of TBR for better support of vehicular network specificities where service degradation or forced call termination may occur owing to frequent handoffs. Detailed simulation results for TBR and VTBR schemes and a comparison with the guard channel scheme are presented. The results show that TBR and VTBR can efficiently improve the flow dropping probability. Copyright © 2008 John Wiley & Sons, Ltd. [source]

Active resource management for the differentiated services environment

Manish Mahajan
This paper presents a mechanism for active resource management (ARM) in a differentiated services environment. While the differentiated services architecture and the bandwidth broker agent provide a mechanism for QoS management through resource reservation, this mechanism is based on a static provisioning of resources. As bandwidth requirement are typically dynamic, such a static reservation approach can either lead to wasted bandwidth or leave applications resource-starved. The active resource management approach presented in this paper addresses this problem by dynamically reallocating resources based on current network state and applications requirements. An implementation and evaluation of ARM using the NS-2 simulation toolkit is also presented.,Copyright © 2004 John Wiley & Sons, Ltd. [source]