Real Network Applications Research Articles

Methodology for Locating Link Count Sensors that Accounts for Reliability of Prior Estimates from Origin–Destination Matrices

The paper proposes an innovative methodology for addressing the issue of the location of optimal link count sections on the basis of the proposition of a reliability measure in which the prior accuracy of the estimate of the origin–destination (O-D) matrix, that is, its statistical distribution rather than its prior punctual estimate, is explicitly considered, together with its posterior distribution conditioned on a given subset of link count locations. The proposed measure, under the mild assumptions of prior normal distribution and through appropriate algebraic rotations of the reference system, is proved not to depend on the unknown values of the counted flows; this is actually the key point that allows for effective implementation of a fast and operational procedure that is based on this mathematical framework. As an example, toy network and real network applications are presented to show how a heterogeneous level of knowledge across O-D pairs may lead to the choice of counting sections different from those resulting from the commonly adopted procedures. The proposed methodology allows for a more effective theoretical interpretation of the phenomenon and leads to an extremely efficient computational procedure—suitable also in real-size networks—whose results outperform those obtained with the methods currently available in the literature.

Heat flux distribution and rectification of complex networks

It was recently found that the heterogeneity of complex networks can enhance transport properties such as epidemic spreading, electric energy transfer, etc. A trivial deduction would be that the presence of hubs in complex networks can also accelerate the heat transfer although no concrete research has been done so far. In the present study, we have studied this problem and have found a surprising answer: the heterogeneity does not favor but prevents the heat transfer. We present a model to study heat conduction in complex networks and find that the network topology greatly affects the heat flux. The heat conduction decreases with the increase of heterogeneity of the network caused by both degree distribution and the clustering coefficient. Its underlying mechanism can be understood by using random matrix theory. Moreover, we also study the rectification effect and find that it is related to the degree difference of the network, and the distance between the source and the sink. These findings may have potential applications in real networks, such as nanotube/nanowire networks and biological networks.

Thread to strand binding of parallel network applications in massive multi-threaded systems

In processors with several levels of hardware resource sharing,like CMPs in which each core is an SMT, the scheduling process becomes more complex than in processors with a single level of resource sharing, such as pure-SMT or pure-CMP processors. Once the operating system selects the set of applications to simultaneously schedule on the processor (workload), each application/thread must be assigned to one of the hardware contexts(strands). We call this last scheduling step the Thread to Strand Binding or TSB. In this paper, we show that the TSB impact on the performance of processors with several levels of shared resources is high. We measure a variation of up to 59% between different TSBs of real multithreaded network applications running on the UltraSPARC T2 processor which has three levels of resource sharing. In our view, this problem is going to be more acute in future multithreaded architectures comprising more cores, more contexts per core, and more levels of resource sharing. We propose a resource-sharing aware TSB algorithm (TSBSched) that significantly facilitates the problem of thread to strand binding for software-pipelined applications, representative of multithreaded network applications. Our systematic approach encapsulates both, the characteristics of multithreaded processors under the study and the structure of the software pipelined applications. Once calibrated for a given processor architecture, our proposal does not require hardware knowledge on the side of the programmer, nor extensive profiling of the application. We validate our algorithm on the UltraSPARC T2 processor running a set of real multithreaded network applications on which we report improvements of up to 46% compared to the current state-of-the-art dynamic schedulers.

A Novel Dynamic Fault Restoration Mechanism Using a Multiple Ring Approach in WDM Mesh Networks

This work describes a distributed fault restoration algorithm, called the Dynamic Multiple Ring Algorithm (DMRA), for application in WDM mesh networks. This study explores the choice of restoration paths and the assignment of fault-tolerant bandwidth when a link, node, or channel failure occurs according to the change in traffic load, number of nodes, and transmission delay including propagation and switching delays. Accordingly, the primary aim of this work is to use networking segments near faults to share the restoration load throughout a mesh network. Each node searches for restoration paths in their near environment using the proposed DMRA. Nodes use distributed control to search for neighboring nodes and to establish the relationship between them to build numerous logical rings. Nodes can also locate faults in the logical rings. These rings establish the restoration paths. The traffic load over failed links or nodes can be diverted to other paths in the networking segments. The cost of the restoration paths is computed at each node based on both the current capacity and the transmission delay. The selected restoration paths are suitable transmission routes in the network neighborhood. Hence, restoration paths can be identified and wavelength assigned quickly according to network bandwidth and traffic load. Simulation results reveal that the proposed method works extremely quickly and has a high success rate. Consequently, it is very useful for applications in real WDM networks, where the status varies from minute to minute.

Minimal Loss Reconfiguration Based on Dynamic Programming Approach: Application to Real Systems

A method for minimal losses in large primary distribution systems by changing topology is presented. The loss minimization in on-line condition in large real distribution systems is based on dynamic programming approach with graph compression technique and radial power flow. The dynamic programming naturally adapts to operational spanning tree configuration of primary networks and it is easy to implement. The principal characteristics of the proposed method are applications to large real systems and applicability to ill-conditioned systems. It considers all load type andit requires very low computational efforts. The model is useful for planning and operating in primary distribution systems. Through different test network applications, the method was validated and compared to other proposed methods. This paper shows two real network applications and the results show how powerful the method is when applied to large systems in very short times.

LAN Emulation over ATM

This paper is aimed at network designers and engineers who are looking at the deployment of ATM networking technology within or alongside an existing LAN infrastructure. The paper describes in detail how an ATM network can be made to emulate an existing Ethernet or Token Ring LAN. The benefit of this scheme is that it allows existing LAN applications to run over an ATM network with no modification, and it provides for straightforward interconnection of existing Ethernet and Token Ring LANs with an ATM network. The paper provides a comprehensive tutorial on the LAN Emulation scheme, which is the subject of standardization by the ATM Forum, together with a discussion of the practical implications of the scheme and its application in real networks.