A resource management architecture over differentiated services domains for guarantee of bandwidth, delay and jitter

Hashmani

2008 6th National Conference on Telecommunication Technologies and 2008 2nd Malaysia Conference on Photonics

2008

There are many fields at where ad hoc networks are used, but applications like mission critical and commercial production or broadcasting which mostly contain real time audio and video streams and require absolute end to end guarantee for QoS. The key attribute these applications for guarantying for QoS is, they need routing path for connectivity like virtual leased line that is very difficult in the scenario of mobile ad hoc networks due to the mobility of nodes and changing topology conditions. With respect to transfer the contents, resources are required to be reserved along the selected path(s).In this regard previously two types of approaches are being used. load balancing in which multiple paths are selected and resources are reserved on those andpackets are sent along all paths and other to use redundant paths but using only one at one time by reserving resources on that. Onfailure offirst one, second is used and tried to reserve resources on that alternate path. In this paper we have introduced entirely a new approach in which we find multiple paths and select best three from all those as primary, secondary and ternary on certain criteria of maximum bandwidth and minimum hopes and these paths must be loopfree and node disjoined.Resources are reserved on both primary and secondary paths and allocate those reserved resources to different categories of traffic according to their priority on the basis ofservice agreement made with network. We have designed the complete mechanism for allocating the resources to different priority traffics and tried to avoid the route breaks.

Section: Related Work and Discussionmentioning

confidence: 99%

Premeditated Policies to Resource Reservation for Prioritized Traffic in Multi Hop Ad Hoc Networks

Hashmani

2008 6th National Conference on Telecommunication Technologies and 2008 2nd Malaysia Conference on Photonics

2008

WIREs Computational Stats

“…The cluster platform varies over many operating systems such as Linux,15,32–35, Solaris,13,36–38 and Windows‐NT 39. Many resource management tools have been proposed as the middleware, such as Maui Scheduler,22 Condor,8,10,12,19,20,26,28,40–42 Portable Batch System (PBS),43 Sun Grid Engine (SGE),37 Load Sharing Facility (LSF),21,44 CODINE,12 Computing Center Software (CCS),45 and CoCheck 35. The work done in this area focused on the aspect of computing power.…”

Section: Discovery and Search In Distributed Grid Systemsmentioning

confidence: 99%

Distributed grid‐type architectures

Alnoshan

Rotenstreich

2010

Research advances in the distributed computing have led to grid computing, which unlike many other areas of distributed computing focuses on large-scale resource sharing and discovery for better performance. Traditionally, grid computing has been performed in a client/server format. However, recent advances in the area of peer-to-peer computing have brought a paradigm shift in the field of distributed systems and have forced researchers in the area of grid computing to discuss peer-to-peer grids. In this article, we will review the work done in the area of grid computing and discuss peer-to-peer grids with a focus on a framework that provides incentive to the participating resources in a peer-to-peer grid environment. The framework provides the basis for defining a utility function that is used by participating nodes to efficiently allocate the resources that are available in a system. We finish by mentioning the evolving area of cloud computing.

“…2). Since its first appearance, our BB has been constantly enhanced to include many new features and accommodate new standards as well [7] [8]. We have performed several QoS experiments deploying our BB over wide area networks (WAN) in Japan.…”

Section: Our Bbmentioning

confidence: 99%

Management and Realization of SLA for Providing Network QoS

Hashmani¹,

Yoshida²,

Ikenaga

et al. 2001

Networking — ICN 2001

Abstract.We consider an SLA (Service Level Agreement) committed between two parties to use the guarantee of QoS provided by a QoS Enabled Network (QEN). QEN can provide guarantee of QoS because a Bandwidth Broker (BB) manages its resources based on a policy. The IETF and the DMTF have done extensive work in this field and have proposed a policy framework in order to store and manage policy and the standardization process is still going on. However, not much work has been done in the field of realization and management of SLA, which is necessary in order to utilize QoS provided by the network layer as perceived by the service layer. In this paper, we propose a new methodology of SLO templates (machine readable description of human readable textual SLA) to negotiate and commit SLA to utilize QoS provided by networks as perceived by businesses like that of a virtual leased line service. We also describe the implementation of our proposed SLO templates in our BB.