دانلود مقاله ISI انگلیسی شماره 7987
ترجمه فارسی عنوان مقاله

الگوریتم هیوریستیک برای حفاظت از بخش مشترک در شبکه های تقسیم چندگانه طول موج با مسیر پشتیبان گیری محدود / طول بخش

عنوان انگلیسی
A heuristic algorithm for shared segment protection in mesh WDM networks with limited backup path/segments length
کد مقاله سال انتشار تعداد صفحات مقاله انگلیسی
7987 2006 17 صفحه PDF
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Computer Communications, Volume 29, Issue 16, 12 October 2006, Pages 3197–3213

ترجمه کلمات کلیدی
- زنده ماندن - الگوریتم های مسیریابی - حفاظت از بخش به اشتراک گذاشته شده - زمان حفاظت سوئیچینگ
کلمات کلیدی انگلیسی
پیش نمایش مقاله
پیش نمایش مقاله  الگوریتم هیوریستیک برای حفاظت از بخش مشترک در شبکه های تقسیم چندگانه طول موج با مسیر پشتیبان گیری محدود / طول بخش

چکیده انگلیسی

This paper investigates the problem of dynamic survivable lightpath provisioning against single-link failure in optical mesh networks employing wavelength-division multiplexing (WDM). We focus on the special problem of provisioning lightpath requests according to their differentiated protection-switching time, since lightpath may have different protection-switching time requirements. We assume that the protection-switching time requirements of connections can be transformed to the hop limits of backup path/segments by using techniques proposed in the literature such as [Y. Luo, N. Ansari, Survivable GMPLS networks with QoS guarantees, IEE Proc. Commun., vol. 152, (4) (2005) 427–431]. We propose a heuristic algorithm, namely Suurballe-based Heuristic Algorithm using Least number of segments for SSP with hop Limit (SHALL), to efficiently solve this problem. We inspect the effects of hop limit on various performance matrices and compare the SHALL approach with three other well-known protection approaches, namely shared path protection (SPP), shared link based protection (LBP) and cascaded diverse routing (CDR). Numerical results demonstrate that the SHALL approach outperforms its counterparts in blocking probability and protection-switching time with mirror decrease of spare capacity efficiency.

مقدمه انگلیسی

Wavelength-routed wavelength division multiplexing (WDM) networks have been considered to be a promising network infrastructure for future backbone transport networks. In such networks, each fiber link offers huge bandwidth capacity to carry user traffic. A single network failure may cause a large amount of data loss in the network, which would largely degrade network performance and even disrupt network services. Thus network survivability is of great importance in such networks. To guarantee network services, the network must incorporate effective protection and restoration capabilities to survive different kinds of network failures (e.g., a fiber cut or a node fault). Although the higher layers (e.g., IP, ATM and SONET) may provide their own protection and restoration mechanisms, it is still attractive to provide protection and restoration capabilities at the optical layer because of a number of advantages, such as fast service recovery, efficient resource utilization, and protocol transparency. Survivability is the ability of the network to withstand equipment and link failures. The main goals of survivable network design are to be able to perform rapid restoration at as small a cost as possible (i.e., using minimum resources). Node equipment failures are typically handled using redundant equipment within the node (including redundant switches). On the other hand, link failures, which are by far the most common failures in optical networks, occur due to backbone accidents. In this paper, we restrict ourselves to the case of link failures. At the same time, there is an increasing importance for service providers to provide guaranteed service in recent years. This entails that survivable routing schemes should not only be both capacity- and computational-efficient, but also minimize the possible restoration time for a specific connection, such that the maximum benefits can be gained in the operation of carrier networks [15]. Segment shared protection (SSP) [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13] and [14] is one of the best approaches to meet the above design requirements, where a working path is divided into a set of possible overlapping active segments (AS), and provide protection for some or all of the links along each AS using a backup segment (BS), which is link/node-disjoint with the AS. Comparing with its counterpart-shared path protection (SPP) [15], [16], [17], [18], [19], [20] and [21], SSP has been reported to achieve a better throughput by maximizing the extent of spare capacity resource sharing. In this paper, we investigate the problem of how to efficiently derive backup segments with limited hop length for a given working path in a dynamic network environment, where connection requests arrive dynamically one after another. In the following, we present the art of state of SSP and motivate our study.

نتیجه گیری انگلیسی

In this paper, we investigate the problem of dynamic survivable lightpath provisioning against single-link failures in optical mesh network employing wavelength-division multiplexing (WDM). We focus on the special problem of provisioning lightpath requests according to their differentiated protection-switching time, since lightpath may have different protection-switching time requirements. We propose a heuristic algorithm, namely Suurballe-based Heuristic Algorithm using Least number of segments for SSP with hop Limit (SHALL), to solve this problem efficiently. We inspect the effects of hop limit and ε on various performance matrixes. We found that the hop limit and ε have great effects on each performance matrix. Specifically, the larger the value of ε, the larger the blocking probability and the longer the average hop number of working path/segments. However, the average hop number of backup path/segments and the average total hop number of working/backup path/segments decrease when ε increases. We also found that the blocking probability decreases with the increase of H, which implies that larger value of H is preferable. We have compared the SHALL approach with three other well-known protection approaches, namely shared path protection (SPP), shared link based protection (LBP) and cascaded diverse routing (CDR). Numerical results demonstrate that the SHALL approach outperforms its counterparts in blocking probability and protection-switching time with mirror decrease of spare capacity efficiency.