论文信息 - Approximating optimal spare capacity allocation by successive survivable routing

2003 - The Kips Transactions:partc

A Study of Efficient Spare Capacity Planning Scheme in Mesh-Based Survivable Fiber-Optic Networks

Due to the development of information technology and widespread use of telecommunications networks, the design of mesh-survivable net works has received considerable attention in recent years. This paper deals with spare capacity planning scheme in mesh-based fiber-optic networks. In this study, a new spare capacity planning scheme is proposed utilizing path restoration with maximal sharing of share capacity that is traced by the spare capacity incremental factor (after this, we called "SCIF"). We compare it with three other spare capacity planning scheme : link capacity of IP (Integer Programming), SLPA(Spare Link Placement Algorithm) and GA(Genetic Algorithm). The approach shows better performance with heuristics algorithm for determining the spare capacity assignment and the computational time is easily controlled allowing the approach to scale to large networks. The major advantages of the new approach are reduction of spare capacity and a polynomial time complexity.omplexity.

2007

Design and analysis of survivable telecom mesh networks

Optical fiber technology employing wavelength-division multiplexing (WDM) has been and continues to be investigated and commercially deployed to meet bandwidth-intensive application demands. A fiber cut may result in huge data and revenue loss. Therefore, it is very desirable to design a survivable telecommunications network. This dissertation investigates design and modeling issues of survivable telecom mesh networks. This dissertation first explores the problem of providing availability guarantees for differentiated services. We present a mathematical model for availabilities of connections with shared-path protection by applying the link-vector technique. Based on the model, we propose an efficient provisioning algorithm to provide differentiated services, either no protection, shared-path protection, or dedicated-path protection to dynamic connection requests in a network. The strategy jointly considers both availability satisfaction and resource optimization. Next, we investigate the problem of backup-bandwidth reprovisioning after network-state updates due to connection arrivals and departures as well as failure occurrences and repairs. Rearranging backup bandwidth periodically for existing connections upon network-state updates helps to reduce resource redundancy and improve the blocking performance. We propose both Integer Linear Program (ILP) formulation and heuristic algorithms. We investigate the effects of different reprovisioning periods and some additional dimensions of backup-bandwidth reprovisioning. We also study the effects of multiple backup paths and link sharing among primary and backup paths for availability-guaranteed and service-differentiated provisioning. We present an availability-analysis model for connections with 1+N (N≥1) dedicated-path protection scheme with k ( k≥1) shared links among primary and backup paths. Based on the model, we propose a provisioning algorithm to explore the effect of link sharing among one primary and N backup paths such that good capacity utilization can be achieved. We develop an accumulated-downtime-aware restoration algorithm for connection recovery against link failures. Upon a link failure occurrence, a faulty connection is switched to its pre-assigned backup route or an alternate restoration path (if resources are available), only when its accumulated downtime plus the link repair time will violate its service-level agreement (SLA) requirement. This dissertation makes important contributions by introducing and studying novel models, algorithms and architectures that will help to improve survivability of next-generation telecom networks.

2011

Performance and memory space optimizations for embedded systems

Embedded systems have three common principles: real-time performance, low power consumption, and low price (limited hardware). Embedded computers use chip multiprocessors (CMPs) to meet these expectations. However, one of the major problems is lack of efficient software support for CMPs; in particular, automated code parallelizers are needed. The aim of this study is to explore various ways to increase performance, as well as reducing resource usage and energy consumption for embedded systems. We use code restructuring, loop scheduling, data transformation, code and data placement, and scratch-pad memory (SPM) management as our tools in different embedded system scenarios. The majority of our work is focused on loop scheduling. Main contributions of our work are: We propose a memory saving strategy that exploits the value locality in array data by storing arrays in a compressed form. Based on the compressed forms of the input arrays, our approach automatically determines the compressed forms of the output arrays and also automatically restructures the code. We propose and evaluate a compiler-directed code scheduling scheme, which considers both parallelism and data locality. It analyzes the code using a locality-parallelism graph representation, and assigns the nodes of this graph to processors. We also introduce an Integer Linear Programming based formulation of the scheduling problem. We propose a compiler-based SPM conscious loop scheduling strategy for array/loop based embedded applications. The method is to distribute loop iterations across parallel processors in an SPM-conscious manner. The compiler identifies potential SPM hits and misses, and distributes loop iterations such that the processors have close execution times. We present an SPM management technique using Markov chain based data access prediction for irregular accesses. We propose a compiler directed integrated code and data placement scheme for 2-D mesh based CMP architectures. Using a Code-Data Affinity Graph (CDAG) to represent the relationship between loop iterations and array data, it assigns the sets of loop iterations to processing cores and sets of data blocks to on-chip memories. We present a memory bank aware dynamic loop scheduling scheme for array-intensive applications. The goal is to minimize the number of memory banks needed for executing the group of loop iterations.

2019 - Biologia

Substantial enhancement of high polymorphic SSR marker development using in silico method from 18 available rice blast fungus genome sequences and its application in genetic diversity assessment

Rice blast fungus, Magnaporthe oryzae, causes rice blast disease leading to the reduction of rice yield. Utilization of rice blast resistant cultivar is one of efficient ways to cope with this disease. The genetic diversity of rice blast is of great importance as it indicates the genotype of rice blast isolates in a population of epidemic area and reveals the dynamics of genotype change over time. Simple sequence repeat (SSR) marker is frequently used for indicating the genetic diversity of several organisms including rice blast fungus. The in silico SSR development method provides the enhancement of polymorphic rate and requires less time and labour. This study aimed to develop in silico SSR marker and study genetic diversity of rice blast isolates collected from Thailand. The 11,072 SSR loci were examined in rice blast reference genome 70–15. The most abundant SSR type and pattern were mono-nucleotide repeat and AGC/CTG motif respectively. Fourteen highly polymorphic SSR loci, identified by in silico from 18 rice blast genome sequences, were selected and specific primer were developed. Ten SSR loci were novel and detected the polymorphism in Thai rice blast isolates, which showed 2–7 alleles per locus and PIC values ranged from 0.07 to 0.56. The genetic relationship analysis showed that rice blast isolates were clustered by collection time and site. Our result indicated that in silico SSR development is a powerful technique to enhance the production of highly polymorphic SSR markers and these newly developed SSR markers were useful for genetic diversity assessment.

2016

Initiating Sustained Silent Reading

a simple objective of any school's reading program is to develop each student's ability to read silently without interruption for a relatively long period of time. Dr. Lyman C. Hunt, Jr. of the University of Vermont designated such a goal as early as 1960 and soon thereafter introduced Uninterrupted Sustained Silent Reading with the acronym USSR. The acronym was to command attention; we found it commanded too much attention so we dropped the U and SSR evolved. Initially, the sustaining power was thought to be a goal reached through months or years of practice. We discovered that the goal can be reached almost instantly if SSR is viewed as a complement of a teaching program. Since SSR takes school time, it supplants some part of each student's existing school program. It may supplant part of a student's reading program, but it is more effective when it becomes a part of each teacher's program. SSR is the drill of silent reading; it is the drill or practice necessary in learning to read, not a total reading program. SSR requires no special machines or materials other than a timer; it requires no special teacher training or expertise.

2001

Surface structure of amorphous PMMA from SPM: auto‐correlation function and fractal analysis

The surface topography of amorphous poly(methyl methacrylate) has been investigated by scanning probe microscopy technique and analysed using an auto-correlation function approach. Spatial correlations in the arrangement of sub-macromolecular fragments on the surface are studied depending on preparation conditions (whether the surface was free, i.e. in contact with air, or confined with the silicon substrate). The correlation length of the roughness distribution on the surface of amorphous polymers corresponds to the internal dimensions of unperturbed polymer chains (the radius of gyration). The scaling behaviour of the roughness is analysed with the Hurst approach, and found to be different for the free surface and for the confined one. As a result of annealing at a temperature above the glass transition temperature of the bulk polymer, both surfaces converge to similar fractal dimensionalities and correlation lengths.

2018 - 2018 10th International Workshop on Resilient Networks Design and Modeling (RNDM)

Survivable Routing in Spectrally-Spatially Flexible Optical Networks with Back-to-Back Regeneration

In this article, we address the problem of survivable routing in spectrally-spatially flexible optical networks (SS-FONs) with dedicated path protection (DPP). The considered SS-FON supports spectral super-channel transmissions, in which flexible signal regeneration is realized with transceivers operating in back-to-back (B2B) configurations. We propose an Adaptive Survivable Routing with Back-to-Back Regeneration (ASRBR) algorithm to realize dynamic routing requests protected with the DPP method. The ASRBR algorithm is aware of limited spectrum and transceivers resources and makes use of the flexibility of the B2B regeneration. With the use of the ASRBR algorithm, we examine potential performance gains in terms of bandwidth blocking probability (BBP) in two regeneration scenarios: a reference one in which the use of regenerators is minimized and the modulation conversion is not allowed and intentional regeneration with possible modulation conversion. We analyze the efficiency of ASRBR for a set of various network scenarios using two representative topologies. Evaluation results show that the proposed method outperforms other reference algorithms. Moreover, the flexible B2B regeneration provides better utilization of both spectrum and transceiver resources, resulting in lower BBP than the minimal regeneration scenario. Finally, we study potential performance gains from applying a squeezed protection approach.

2006 - IEEE Globecom 2006

OPN02-2: Inter-Group Shared Protection (I-GSP): A Scalable Solution for Survivable WDM Networks

The past studies for survivable routing suffers from the scalability problem when the number of nodes or connection requests grows in the network. In this proposal, a novel path based shared protection framework namely Inter-Group Shared protection (I-GSP) is developed such that the traffic matrix can be divided into multiple protection groups (PGs) based on specific grouping policy. This novel scheme not only overcomes the scalability problem but also provides an upper bound on the affected working paths in case of link failure in the network. Experiment results show that I-GSP based integer linear programming model solves the networks in a reasonable amount of time for which a regular integer linear programming formulation becomes computationally intractable. For most of the cases the performance gap between the optimal solution and the proposed I-GSP ranges between (2-16)%. The proposed optimization model yields a scalable and near-optimal solution for the capacity planning in the survivable optical networks.

2013 - International Journal of Electrical Power & Energy Systems

Sub-Synchronous Resonance damping via Doubly Fed Induction Generator

Abstract Interests of low frequency oscillation damping with the Doubly Fed Induction Generator (DFIG) has been increased recently. In addition, it is widely accepted that, the Sub-Synchronous Resonance (SSR) is another unfavorable dynamical phenomenon in power systems. In this study, the implementation of the DFIG in SSR mitigation will be investigated. The IEEE second benchmark model aggregated with a DFIG based wind farm is employed as the case study. Two different control methodologies are proposed: a Fuzzy Logic Damping Controller (FLDC) and a conventional damping controller (CDC) that are added to the main control loop of the DFIG in order to damp the SSR. To validate the results of SSR suppression via DFIG, several case studies are introduced based on changing the operation point of the system and duration of the fault. It has been shown that the DFIG can damp the SSR through both proposed methodologies, but when the system operating point or fault duration is changed, the FLDC can present a cost-effective solution for SSR damping. The Fast Fourier Transform (FFT), simulation results, and a method based on performance index (PI) are conducted to compare two controllers through various cases in Matlab.

2006 - IEEE Globecom 2006

OPN08-03: Backup Path Re-optimizations for Shared Path Protection in Multi-domain Networks

Within the context of dynamic routing models for shared path protection in multi-domain networks, we propose a backup path re-optimization phase with possible rerouting of the existing backup paths in order to increase the bandwidth sharing among them while minimizing the network backup cost. The re- optimization phase is activated periodically or when routing a new connection fails because of insufficient capacity. Three re- optimization models are discussed: i) Global rerouting where the re-optimization is performed once for the entire network; ii) Local rerouting where the re-optimization is serially performed on one domain at a time or on selected domains, and iii) Local rerouting with least effort, i.e., where the smallest possible number of backup path reroutings is performed in order to be able to handle new connection requests. The first model offers the best resource savings while the two others are more scalable in multi-domain networks. Comparative performance of the three models are conducted and numerical results are presented.

2008 - IEEE/ACM Transactions on Networking

Maximizing Restorable Throughput in MPLS Networks

MPLS recovery mechanisms are increasing in popularity because they can guarantee fast restoration and high QoS assurance. Their main advantage is that their backup paths are established in advance, before a failure event takes place. Most research on the establishment of primary and backup paths has focused on minimizing the added capacity required by the backup paths in the network. However, this so-called Spare Capacity Allocation (SCA) metric is less practical for network operators who have a fixed capacitated network and want to maximize their revenues. In this paper, we present a comprehensive study on restorable throughput maximization in MPLS networks. We present the first polynomial-time algorithms for the splittable version of the problem. For the unsplittable version, we provide a lower bound for the approximation ratio and propose an approximation algorithm with an almost identical bound. We present an efficient heuristic which is shown to have excellent performance. One of our most important conclusions is that when one seeks to maximize revenue, local recovery should be the recovery scheme of choice.

2003

Spare Capacity Allocation

This chapter introduces spare capacity allocation schemes performed either off-line for a static deployment, or on-line during a time interval between two consecutive network events (i.e., a connection setup or tear-down). As survivability becomes one of the most important issues in the design of control and management of WDM networks, development of protection and restoration schemes became critical to network applicability and/or revenue-making capability. As mentioned in Chapter 6, pre-planning or pre-configuration of spare capacity for some working paths with stringent quality-of-service requirements, are the most commonly seen approaches for achieving this goal. However, if the allocation of spare capacity cannot be both capacity- and computation-efficient, the applicability of the algorithms may be limited, and the network cost for control and management increased. Therefore, we need to develop a spare capacity allocation scheme which meets the requirements of computation-efficiency, capacity-efficiency, class of service, and adaptability to traffic variation.

2006

Backup Capacity Requirements for MPLS Fast Reroute

MPLS fast reroute (MPLS-FRR) mechanisms deviate the traffic in case of network failures at the router closest to the outage location to achieve an extremely fast reaction time. We review and compare the one-to-one backup and the facility backup concept that are options for MPLS-FRR to deviate the traffic via a detour or a bypass, respectively. Basically, the backup paths can take the shortest path that avoids the outage location from the point of local repair to the tail-end router or to the merge point. We then evaluate the backup capacity requirements and the configuration overhead in terms of the number of backup paths per primary path in a parametric study depending on the network characteristics. While the facility backup concept imposes clearly less configuration overhead than the one-to-one backup, its standard path layout requires more capacity. This can be reduced by a simple modification.

2007 - 2007 2nd IEEE/IFIP International Workshop on Broadband Convergence Networks

Heuristic Survivable Routing Algorithm for Multiple Failures in WDM Networks

In this paper, we study the problem of multiple failures in WDM networks and propose a new heuristic algorithm called shared multi-sub-backup-paths reprovisioning (SMR) to improve the survivable performance. Compared with previous algorithms, the survivable performance of SMR in multiple failures can be significantly improved.

2011 - Comput. Commun.

Algebraic connectivity metric for spare capacity allocation problem in survivable networks

For studying survivability of telecommunication networks, one should be able to differentiate topologies of networks by means of a robust numerical measure that can characterize the degree of immunity of a given network to possible failures of its elements. An ideal metric should be also sensitive to such topological features as the existence of nodes or links whose failures are catastrophic in that they lead to disintegration of a given network structure. In this paper, we show that the algebraic connectivity, adopted from spectral graph theory, namely the second smallest eigenvalue of the Laplacian matrix of the network topology, is a numerical index that characterizes a network's survivability better than the average node degree that has been traditionally used for this purpose. This proposition is validated by extensive studies involving solutions of the spare capacity allocation problem for a variety of networks.

2003 - GECCO

Optimization of Spare Capacity in Survivable WDM Networks

A network with restoration capability requires spare capacity to be used in the case of failure. Optimization of spare capacity is to find the minimum amount of spare capacity for the network to survive from network component failures. In this paper, this problem is investigated for wavelength division multiplexing (WDM) mesh networks without wavelength conversion. We propose a hybrid genetic algorithm approach (GA) for the problem. Simulated Annealing (SA) and Tabu Search (TS) are also applied to this problem for comparison purpose. Simulation results show very favorable results for the Genetic Algorithm approach.

2010 - 2010 IEEE Global Telecommunications Conference GLOBECOM 2010

A Hybrid Protection/Restoration Scheme for Two-Link Failure in WDM Mesh Networks

We consider the problem of handling two-link failures in WDM mesh networks. Existing schemes either prereserve two backup paths for each demand or compute new backup paths for unprotected demands after the first link failure occurs. Both approaches require a large amount of backup capacity. In this paper, we propose a capacity efficient hybrid protection/restoration scheme for handling two-link failures. The protection component reserves backup capacity intelligently to ensure the majority of the affected demands can be restored using the pre-planned backup paths upon a two-link failure. For the rest of the affected demands, our restoration component dynamically finds new backup paths to quickly restore them. The capacity efficiency and protection/restoration capability of our hybrid scheme are demonstrated through numerical results.

2002 - Proceedings ISCC 2002 Seventh International Symposium on Computers and Communications

Spare capacity planning using survivable alternate routing for long-haul WDM networks

Spare capacity planning is designed to find the minimum amount of spare capacity to be allocated throughout a network so that the network can survive from network component failures. In this paper, the spare capacity planning problem is investigated for long-haul wavelength division multiplexing (WDM) networks. A three-step method is developed for solving the problem. First, heuristic approaches are used to select both candidate working routes and protection routes in order to achieve approximate optimal performance while maintaining a computational feasibility. Second, traffic requests are distributed on the obtained candidate working and protection routes optimally using genetic algorithms (GA). Finally wavelengths are assigned to working lightpaths and shared protection lightpaths. The major advantage of the new approach is the ability to incorporate nonlinear constraints and nonlinear cost functions into the GA, which are introduced by sharing protection links between shared risk link groups (SRLG). Moreover, by considering SRLG constraints in the spare capacity planning phase, wavelengths can be allocated to each shared protection route before failures happen, so that shorter restoration latency can be achieved. Numerical results illustrate that the proposed approach is more cost-effective than the single-path protection method.

2006 - Journal of Combinatorial Optimization

Finding disjoint paths with related path costs

We consider routing in survivable networks that provide protection against node or link failures. In these networks resilience against failures is provided by routing connections on pairs of disjoint paths called primary and backup paths. The primary path of a connection carries its traffic under normal circumstances and in the eventuality of a network failure effecting the primary path the connection traffic (all or some portion of it) is rerouted over its backup path. In an online setting as connection requests arrive a pair of disjoint primary and backup paths of least total cost (under some link cost metric) are selected to route the connections. In many situations the cost metric used for the primary path differs from the cost metric used for the backup path. Also in many realistic settings these two cost metrics are related to each other.In this paper we study the problem of finding a pair of edge or node disjoint paths of least total cost where the cost of the primary path is the total cost of its links while the cost for the backup path is α times the sum of the cost of its links, for some given α < 1. We show that the problem is hard to approximate to within a factor $$\Omega((\frac{1}{\alpha})^{1-\epsilon})$$ for any positive $$\epsilon \le 1$$. In addition we show that the problem is complete for a set of hard to approximate problems. On the positive side we show that a simple algorithm achieves an approximation ratio of $$O(\frac{1}{\alpha})$$ for the problem.

2014 - IEEE/OSA Journal of Optical Communications and Networking

Optimal design for shared backup path protected elastic optical networks under single-link failure

This paper considers the network protection technique of shared backup path protection (SBPP) in comparison with 1 + 1 path protection for elastic optical networks. We develop integer linear programming (ILP) models to minimize both the required spare capacity and the maximum number of link frequency slots (FSs) used. We consider transponder tunability that corresponds to the condition of whether or not the same set of FSs is required to be used for both the working and protection lightpaths. We also apply the bandwidth squeezed restoration technique to obtain the maximum restoration levels for the affected service flows, subject to a limited FS capacity on each fiber link. Our studies show that the proposed SBPP technique requires much lower spare capacity compared to the traditional 1 + 1 path protection approach. The flexibility of allowing the working and protection lightpaths to use different sets of FSs (i.e., with full transponder tunability) has the advantage of reducing both the number of FSs needed and the spare capacity redundancy required.

Approximating optimal spare capacity allocation by successive survivable routing

Virtual Infrastructure Design for Surviving Physical Link Failures

Topological design of survivable wireless access networks

Efficient Shared Protection Network Design Algorithm that Iterates Path Relocation with New Resource Utilization Metrics

Efficient iterative design algorithm for enhancing resource utilization of shared protected optical path networks

A Delay-Constrained Shared Mesh Restoration Scheme

Traffic grooming for survivable WDM networks - shared protection

LSP partial spatial-protection in MPLS over WDM optical networks

Analysis of resource sharing in transparent networks

Network architecture for joint failure recovery and traffic engineering

Protection in survivable WDM grooming network.

Availability-Aware Spare Capacity Allocation with Partially Protected Rings

Providing resilient quality of service connections in provider-based virtual private networks

A robust optimization approach to network design

A Study on the Design of Survivable Optical Virtual Private Networks (O-VPN)

Survivable routing and wavelength assignment (RWA) in optical virtual private networks (O-VPNs)

Design of Survivable Optical Virtual Private Networks ( O-VPNs )

Impact of spare capacity reuse on multi-class lightpath survivability differentiation

Availability Evaluation in Shared-Path-Protected WDM Networks with Startup-Failure-Driven Backup Path Reprovisioning

Wavelength overprovisioning strategies for enhanced optical path restoration

Distributed adaptive diverse routing for voice-over-IP in service overlay networks