Analyzing the Vulnerability of Superpeer Networks Against Attack B. Mitra (Dept. of CSE, IIT Kharagpur, India), F. Peruani(ZIH, Technical University of Dresden, Germany), S. Ghose, N. Ganguly(Dept. of CSE, IIT Kharagpur, India) Junction Outline • • • • Problem Definition Environment Definition Development of the analytical framework Stability of Superpeer Networks against Attack Outline • • • • Problem Definition Environment Definition Development of the analytical framework Stability of Superpeer Networks against Attack Problem Definition • P2P network architecture – All peers act as both clients and servers – No centralized data source – File sharing and other applications like IP telephony, distributed storage, publish subscribe system etc Node Node Node Internet Node Node Problem Definition • Overlay network – – – – An overlay network is built on top of physical network Nodes are connected by virtual or logical links Underlying physical network becomes unimportant Interested in the complex graph structure of overlay Problem Definition • Dynamicity of overlay networks – Peers in the p2p system leave network randomly without any central coordination – Important peers are targeted for attack • DoS attack drown important nodes in fastidious computation – Fail to provide services to other peers • Importance of a node is defined by centrality measures – Like degree centrality, betweenness centraltiy etc • Makes overlay structures highly dynamic in nature • Frequently it partitions the network into smaller fragments • Communication between peers become impossible Problem Definition • Investigating stability of the networks against the churn and attack Network Topology + Attack = How (long) stable • Developing an analytical framework • Examining the impact of different structural parameters upon stability – Peer contribution – degree of peers, superpeers – their individual fractions • Modeling of – Overlay topologies (pure p2p networks, superpeer networks, hybrid networks) – Various kinds of attacks • Defining stability metric • Validation through simulation Outline • Problem Definition • Environment Definition – Modeling superpeer network – Different kind of attack models – Stability metric • Development of the analytical framework • Stability of Superpeer Networks against Attack Environment Definition • Modeling superpeer networks – Simple model : strict bimodal structure • A large fraction (r) of peer nodes with small degree kl • Few superpeer nodes (1-r) with high degree km pk 0, pk 0, if k = kl, km otherwise pkl = r and pkm = 1-r Environment Definition • Different kinds of attack models – Deterministic attack • Nodes having high degrees are progressively removed • qk : the probability that a node of degree k survives after attack • qk = 0, when k > kmax 0 < qk < 1, when k = kmax qk = 1, when k < kmax – Degree dependent attack • Nodes having higher degrees are more likely to be removed • Probability of removal of a node having degree k is proportional to kr where r > 0 is a real number r k • With proper normalization f k , C is a normalizing constant C • The fraction of nodes having degreerk which survives after this kind of attack is qk 1 f k 1 k C Environment Definition • Stability metric – Percolation threshold : • disintegrates the network into large number of small, disconnected components by removing certain fraction of nodes (fc) • Higher values indicate greater stability against attack Stability Matric • Percolation Threshold Initially all the nodes in the network are connected Forms a single giant component Size of the giant component is the order of the network size Nodes in the network are connected and form a single component Giant component carries the structural properties of the entire network Stability Matric • Percolation Threshold f fraction of nodes removed Initial single connected component Giant component still exists Stability Metric • Percolation Threshold f fraction of nodes removed Initial single connected component Giant component still exists fc fraction of nodes removed The entire graph breaks into smaller fragments Therefore fc =1-qc becomes the percolation threshold Percolation Threshold • Remove a fraction of nodes ft from the network in step t and check whether reach the percolation point • CSt (s) sns / s sns – s : size of the components formed – ns : number of componets of size s – CSt(s) : the normalized component size distribution at step t Initial : only single giant component of size 500 Intermediate: Bimodal character (a large component along with a set of small components) Percolation point(tn) percolation threshold (ftn) monotonically decreasing function Outline • Problem Definition • Environment Definition • Development of the analytical framework – Generating function • Stability of Superpeer Networks against Attack Development of the analytical framework • Generating Function: – Formal power series whose coefficients encode information a x Here (a0 , a1 , a2 ,.....) encode information about a sequence – Used to understand different properties of the graph – G ( x) p x generates probability distribution of the vertex degrees. Edge – Average degree z k G0 ' (1) P ( x ) a0 a1 x a2 x 2 a3 x ......... 3 k 0 k 0 k Vertex Degree = 5 k k 0 k Development of the analytical framework – pk .qk specifies the probability of a node having degree k to be present in the network after (1-qk) fraction of nodes removed. – F0 ( x) pk qk x k becomes the corresponding generating k 0 function . (1-qk) fraction of nodes removed – Distribution of the outgoing edges of first neighbor of a randomly chosen node kp q x F ( x) kp k k k 1 k k k 1 F0 ( x) z Random node First neighbor Development of the analytical framework – H1(x) generates the distribution of the size of the components that are reached through random edge – H1(x) satisfies the following condition F1(x) : the probability of finding a node following a random edge => 1 - F1(x) : the probability of following a randomly chosen edge that leads to a zero size component. The rest condition reached through random edge, which satisfies a Self-consistency condition. Development of the analytical framework – H 0 ( x) generates distribution for the component size to which a randomly selected node belongs to – Average size of the components F 0 (1) F1 (1) H 0 (1) F0 (1) 1 F1 (1) – Average component size becomes infinity when 1 F1(1) 0 – theoretically ‘infinite’ size graph reduces to the ‘finite’ size components Development of the analytical framework – Average component size becomes infinity when 1 F1(1) 0 – With the help of generating function, we derive the following critical condition for the stability of giant component kpk (kqk qk 1) 0 k 0 Degree distribution Peer dynamics – The critical condition is applicable • For any kind of topology (modeled by pk) • Undergoing any kind of dynamics (modeled by 1-qk) Outline • • • • Problem Definition Environment Definition Development of the analytical framework Stability of Superpeer Networks against Attack – Simulation result Stability of Superpeer Networks against Attack • Theoretically derived results & simulation – Deterministic attack – Degree dependent attack • Network Generation – Represented by a simple undirected graph – Bimodal degree distribution – Graphs with 5000 nodes Undirected Directed graph An undirected arc is an edge that has no arrow. Both ends of an undirected arc are equivalent--there is no head or tail. Deterministic Attack • Two cases may arise in the deterministic attack – 1. The removal of a fraction of superpeers is sufficient to disintegrate the network – 2. The removal of all the superpeers is not sufficient to disintegrate the network. Therefore we need to remove some of the peer nodes along with the superpeers. Recall : when F1 ' (1) 1 , the critical condition for the stability k (k 1) p k qk k k (k 1) p q k k kl , k m k kl k k (k 1) p q k km k k k Deterministic Attack • Case 1: k (k 1) pk qk k (k 1) pk qk k kl k km – fsp : thek critical fraction of superpeer nodes, removal of which disintegrates k (kthe 1giant ) pk (1component fp) k – qk = 1 k k l for k = kl qk = 1 – fsp for k = km k f p 1 kl (kl 1) pkl kf(tar (1 r ) k ( k 1) p q k 1rf ) pp q k kl k k k km k k k • Case 2: k (k 1) p k (k 1) p (1 f ) k k sp k k kl k removed km – fp : fraction of peer to be along with all the superpeers to breack down the betwork k kl ( kl 1) pkl 1 k = k – qk = 1 - fp f sp for kl m ( k m 1) pk m qk = 0 for k = km f tar (1 r ) f sp Deterministic Attack • Parameter degree kl=1,2,3, –Peer Average degree <k> = 10 the removal of only a –fraction Superpeer degree km = 50 of superpeers fraction of peers is reqired breakdown of –causes Increase the peer degree kl gradually A(the peer fraction to be removed. the network changes accordingly) and observe the change in the The high degree peers percolation threshold ftar The increase of peer degree from 1 to 2 and 3 further reduces the fraction of superpeers in the network It is not large enough to form effective connections within themselves connect among themselves and they are not entirely dependent on superpeers for connectivity. The steep increase of stability with peer degree > 5 Deterministic Attack • For kl=1, 3, ftar gradually reduces, since increase in peer contribution Peer contribution: decreases superpeer contribution, –it decreases controlsstability the total bandwidth contributed by the peers which of these For kl=5, at Prc=0.3, a networks also. determines the amount influence superpeer nodes exerts fraction ofof peers is on the network required to be removed to disintegrate the networks. For kl =5, peers are strongly Prc<0.2 does not have – two factors: peer degree & fraction ofconnected peers inamong the network any impact upon the themselves, stability of the network no mater what peer degree is. hence stability is more rK l PrC , where k rkl (1dependent r )km on peer k contribution. Peer degree kl=1 can be disintegrated without attacking peers at all The impact of high degree peers upon the stability of the network becomes more eminent as peer contribution Prc > 0.5. Degree Dependent Attack • Probability of a node of degree k is directly proportional to kγ where γ > 0 is a real number. – Probability of survival of a node having degree k after a degree k dependent attack is qk 1 f k 1 C – Critical condition for the stability of the giant component : k (k 1) p q k k kl , k m k apply : pkl r rk l 1 k km k k m kl , pk m 1 r k kl k m kl (kl 1) (1 r )k m 1 (k m 1) C ( k (k m kl ) k m 2 k ) Degree Dependent Attack • Probability of removal of a node is directly proportional to its degree, hence C km k f • Minimum value k C • This yields an inequality 1 1 rkl (kl 1) (1 r )km (km 1) km (k (km kl ) km 2k ) – The solution set of the above inequality can be bd ( 0 • either bounded c c ) • either bounded (0 c ) Degree Dependent Attack rkl 1 (kl 1) (1 r )km 1 (km 1) C(k (km kl ) km 2k ) – Obtaining minimum value of C, each γc results in the corresponding normalizing constant rk l c (kl 1) (1 r )k m c 1 (k m 1) C c k ( k m kl ) k m 2 k percolation threshold becomes f c rf p c (1 r ) f sp c kl c km c r (1 r ) C c C c Degree Dependent Attack Case to 2 ofone deterministic attack • The breakdown of the network can be due of the three situations and reasons noted below: – 1: The removal of all the superpeers along with a fraction of peers. bd • Networks having a bounded solution set Src where 0 c c exhibit this kind of behavior at the maximum value of the bd solution c c . cb d • Here the fraction of superpeers removed becomes f sp = 1 k • and fraction of peers removed f p l C – 2: The removal of only a fraction of superpeers. • Some networks have an open solution set Src where 0 c • At c , f p c converges to 0 and f spc converges to some x where 0<x<1. – 3: The removal of some fraction of both superpeers and peers. • Intermediate critical exponents c S c signifies the fractional removal of both peers and superpeers. bd c bd c bd c Degree Dependent Attack • Two superpeer degrees km=25, 50 fixed average degree <k> = 10 • Behavior of peer contribution Prc due to the change in peer degree kl In order to keep the average degree and peer constant, the network with higher superpeer degree results higher fraction of peer which increases the peer contribution. Degree Depend Attack • Behavior of boundary critical exponent due to the change in peer degree Γcbd remains ubounded : peer degree kl < 4 with superpeer degree km = 25 Case 1 of deterministic attack Γcbd remains unbounded : peer degree kl < 3 with superpeer degree km = 50 Removal of only a fraction of superpeers disintegrate these networks: the low peer degree -> low peer contribution -> high superpeer contribution Degree Depend Attack • Fraction of peers and superpeers required to be removed to breakdown the network and its impact upon percolation threshold fc. Peer contribution has profound impact on the stability of the network specially with the networks having high peer degree kl. The gradual increase in peer degree increases the peer contribution -> the higher peer contribution ensures the necessity to remove a fraction of them to breakdown the network. Degree Depend Attack • Case study 1: The removal of all the superpeers along with a fraction of peers. Peer degrees kl =3,4; Average degree <k>=5 Kl=4, spth=4.1 Kl=3, th=1.9 2. Impactspupon the fraction of peers removed: *recall : two factors f c1(c2 ) 1( 2 ) f p1c( 2 ) (1 r1( 2 ) ) bd bd f c c r1 f p1c r2 f p2c (1 r1 ) (1 r2 ) bd bd bd ( rf p c ) (1 r ) bd Depending upon the weightage of influence, bd with respect to (b) Fraction of peers and superpeers (a) Behavior of γ cor increases slowly fp γcbd either decreases Impact the fraction required to be removed to breakdown the 1. change in upon superpeer fractionof peers removed: when the *The fraction of superpeers is lass than bd increase of superpeer fraction slowly increases γ c the network and its impact upon spth. γcbd *Which in turn gradually decreases the percolation fraction of peers removed threshold fc fp *higher degree peers -> higher values of fp γcbd to removed Degree Depend Attack *The fraction of peers removed gradually decreases the 2: increase of critical of only a fraction of superpeers. • Case with study The removal exponent γc which in turn decreases the – Superpeers degree km = 25, average degree <k> = 5, peer degree kl =2 value of fcrc. – Initially remove a fraction of superpeers fsprc and then start removing f peers gradually *As c , p 0 with f sp x c c where(0<x<1) and eventually reach some steady value. *removal of only a fraction of superpeers is sufficient to any network with peer degree kl =1, 2, irrespective of superpeer degree and its fraction since the solution set Src becomes unbounded. γ Degree Depend Attack • Case study 3: The removal of some fraction of both superpeers and peers. – Superpeer degree km=5, average degree <k>=5, peer degree kl=3 Removal of any combination of (fprc, fsprc) where 0<rc<rcbd, results in the breakdown of the network. γcbd = 1.171 Tea Time coming soon