Efficient Integration of Logical Topology Design and Survivable Traffic Grooming for Throughput Enhancement in WDM Optical Networks

Optical networks that employ traffic grooming, merge several low-speed traffic streams onto a high-speed lightpath. A fiber cut (or a link failure) in such networks can disrupt a large number of lightpaths and can cause a huge amount of data loss. Designing survivability schemes to make the network withstand a link failure is thus vital. In this paper we address the problem of survivable traffic grooming in a static scenario. In a static scenario the primary objective is to maximize throughput with the available network resources. Since the problem is NP-Complete, we propose an efficient heuristic TATG (Throughput Aware Traffic Grooming) for the problem. TATG employs an efficient approach of integrating logical topology design and survivable traffic grooming with heuristics at sub-problem level. Our approach ensures that the logical topology carrying the traffic always stay connected in the event of a link failure. Time complexity analysis shows that our heuristic runs in polynomial time. We study two separate forms of survivability in heuristics TATG-SC (TATG -Survivability at Connection) and TATG-SL (TATG-Survivability at Lightpath) with three categories of traffic demand. Performance comparison with a well known heuristic shows that the proposed approach provides much better throughput performance when resources are scarce. The average percentage decrease in request blocking over all experiments obtained using TATG-SC and TATG-SL is 54.87% and 59.57% respectively.