Availability analysis of span-restorable mesh networks

摘要

The most common aim in designing a survivable network is tonachieve restorability against all single span failures, with a minimalninvestment in spare capacity. This leaves dual-failure situations as thenmain factor to consider in quantifying how the availability of servicesnbenefit from the investment in restorability. We approach the questionnin part with a theoretical framework and in part with a series ofncomputational routing trials. The computational part of the analysisnincludes all details of graph topology, capacity distribution, and thendetails of the restoration process, effects that were generally subjectnto significant approximations in prior work. The main finding is that anspan-restorable mesh network can be extremely robust under dual-failurenevents against which they are not specifically designed. In anmodular-capacity environment, an adaptive restoration process was foundnto restore as much as 95% of failed capacity on average over allndual-failure scenarios, even though the network was designed withnminimal spare capacity to assure only single-failure restorability. Thenresults also imply that for a priority service class, mesh networksncould provide even higher availability than dedicated 1+1 APS. This isnbecause there are almost no dual-failure scenarios for which somenpartial restoration level is not possible, whereas with 1+1 APS (ornrings) there are an assured number of dual-failure scenarios for whichnthe path restorability is zero. Results suggest conservatively that 20%nor more of the paths in a mesh network could enjoy this ultra-highnavailability service by assigning fractional recovery capacitynpreferentially to those paths upon a dual failure scenario