How non-uniform tolerance parameter strategy changes the response of scale-free networks to failures

Abstract.In this paper, we introduce a non-uniform tolerance parameter (TP) strategy (the tolerance parameter is characterized by the proportion between the unused capacity and the capacity of a vertex) and study how the non-uniform TP strategy influences the response of scale-free (SF) networks to cascading failures. Different from constant TP in previous work of Motter and Lai (ML), the TP in the proposed strategy scales as a power-law function of vertex degree with an exponent b. The simulations show that under low construction costs D, when b>0 the tolerance of SF networks can be greatly improved, especially at moderate values of b; When b<0 the tolerance gets worse, compared with the case of constant TP in the ML model. While for high D the tolerance declines slightly with the b, namely b<0 is helpful to the tolerance, and b>0 is harmful. Because for smaller b the cascade of the network is mainly induced by failures of most high-degree vertices; while for larger b, the cascade attributes to damage of most low-degree vertices. Furthermore, we find that the non-uniform TP strategy can cause changes of the structure and the load-degree correlation in the network after the cascade. These results might give insights for the design of both network capacity to improve network robustness under limitation of small cost, and for the design of strategies to defend cascading failures of networks.

[1]  S. Havlin,et al.  Breakdown of the internet under intentional attack. , 2000, Physical review letters.

[2]  Cohen,et al.  Resilience of the internet to random breakdowns , 2000, Physical review letters.

[3]  Reuven Cohen,et al.  Stability and topology of scale-free networks under attack and defense strategies. , 2005, Physical review letters.

[4]  Albert-László Barabási,et al.  Statistical mechanics of complex networks , 2001, ArXiv.

[5]  Adilson E Motter.  Cascade control and defense in complex networks. , 2004, Physical review letters.

[6]  Réka Albert,et al.  Structural vulnerability of the North American power grid. , 2004, Physical review. E, Statistical, nonlinear, and soft matter physics.

[7]  Reuven Cohen,et al.  Tolerance of scale-free networks: from friendly to intentional attack strategies , 2004 .

[8]  Albert,et al.  Emergence of scaling in random networks , 1999, Science.

[9]  Petter Holme.  Edge overload breakdown in evolving networks. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[10]  Massimo Marchiori,et al.  Model for cascading failures in complex networks. , 2004, Physical review. E, Statistical, nonlinear, and soft matter physics.

[11]  Y. Hayashi,et al.  Improvement of the robustness on geographical networks by adding shortcuts , 2006, physics/0612065.

[12]  S. N. Dorogovtsev,et al.  Structure of growing networks with preferential linking. , 2000, Physical review letters.

[13]  Beom Jun Kim,et al.  Growing scale-free networks with tunable clustering. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[14]  K. Goh,et al.  Universal behavior of load distribution in scale-free networks. , 2001, Physical review letters.

[15]  Albert-László Barabási,et al.  Error and attack tolerance of complex networks , 2000, Nature.

[16]  S. N. Dorogovtsev,et al.  Evolution of networks , 2001, cond-mat/0106144.

[17]  R. Pastor-Satorras,et al.  Critical load and congestion instabilities in scale-free networks , 2003 .

[18]  Ying-Cheng Lai,et al.  Attack vulnerability of scale-free networks due to cascading breakdown. , 2004, Physical review. E, Statistical, nonlinear, and soft matter physics.

[19]  Adilson E Motter,et al.  Cascade-based attacks on complex networks. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.