2 Push and Pull Source Sink Push

Size: px
Start display at page:

Download "2 Push and Pull Source Sink Push"

Transcription

1 1 Plaing Push vs Pull: Models and Algorithms for Disseminating Dnamic Data in Networks. R.C. Chakinala, A. Kumarasubramanian, Kofi A. Laing R. Manokaran, C. Pandu Rangan, R. Rajaraman

2 Push and Pull 2 Source Sink

3 Push and Pull 2 Push Source Sink

4 Push and Pull 2 Push Source push Sink

5 Push and Pull 2 Push Source push Sink Pull

6 Push and Pull 2 Push Source push Sink Pull quer

7 Push and Pull 2 Push Source push Sink response Pull quer

8 Push and Pull 2 Push Source push Sink response Pull quer Mied Store

9 Push and Pull 2 Push Source push Sink response Pull quer Mied Store

10 Push and Pull 2 Push Source push Sink response Pull quer Mied Store

11 Push and Pull 2 Push Source push Sink response Pull quer Mied Store

12 A Simple Eample Using average source and sink frequencies. 3 Sources Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

13 A Simple Eample Using average source and sink frequencies. 3 Sources Push Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

14 A Simple Eample Using average source and sink frequencies. 3 Sources Push Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

15 A Simple Eample Using average source and sink frequencies. 3 Sources Push Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

16 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

17 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

18 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

19 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

20 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

21 A Simple Eample Using average source and sink frequencies. 3 Sources Pull Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

22 A Simple Eample Using average source and sink frequencies. 3 Sources Mied Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

23 A Simple Eample Using average source and sink frequencies. 3 Sources Mied Store Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

24 A Simple Eample Using average source and sink frequencies. 3 Sources Mied Store Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

25 A Simple Eample Using average source and sink frequencies. 3 Sources Mied Store Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

26 A Simple Eample Using average source and sink frequencies. 3 Sources Mied Store Sink Aggregation No Aggregation Push 4 4 Pull 6 7 Mied Mied is Best 2 < 2 < 4 3 < 2 < 4

27 General Problem High Level 4 INPUTS: Graph G = (V, E) with: cost of updating set of stores: SetC : V Powerset(V ) R + Source Set P V, Sink Set Q V For ever source i P, a source frequenc p i For ever sink j Q, a sink frequenc q j For ever sink j Q, an interest set I j

28 General Problem High Level 4 INPUTS: Graph G = (V, E) with: cost of updating set of stores: SetC : V Powerset(V ) R + Source Set P V, Sink Set Q V For ever source i P, a source frequenc p i For ever sink j Q, a sink frequenc q j For ever sink j Q, an interest set I j OUTPUTS: For ever source i P, a Push set P i For ever sink j Q, a Pull Set Q j Intersection requirement: i I j P i Qj. MINIMIZE: total cost of push-updates, queries and responses: i Pp i SetC(i, P i )+ j Qq j SetC(j, Q j )+ j Q q j RespC(j)

29 Routing Cost Models 5 Multicast Cost Eample Definition Multicast 3 Steiner tree cost Unicast 4 Sum of path costs (non-)metric Distance function Broadcast Model 5 Breadth first tree cost to depth r

30 Routing Cost Models 5 Unicast Cost Eample Definition Multicast 3 Steiner tree cost Unicast 4 Sum of path costs (non-)metric Distance function Broadcast Model 5 Breadth first tree cost to depth r

31 Routing Cost Models 5 Controlled Broadcast Cost Eample Definition Multicast 3 Steiner tree cost Unicast 4 Sum of path costs (non-)metric Distance function Broadcast Model 5 Breadth first tree cost to depth r

32 Related Work 6 FeedTree: RSS via P2P Multicast, [Sandler et al., IPTPS 05] Web Caching applications Combs, Needles and Hastacks Paper, [Liu et al. SENSYS 04] Data Gerrmandering, [Bagchi et al. T.A. TKDE] Minimum Cost 2-spanners: [Dodis & Khanna STOC 99] and [Kortsarz & Peleg SICOMP 98] Multicommodit facilit location, [Ravi & Sinha SODA 04] Classical Theor Problems Facilit Location Steiner Tree (including Group Steiner Tree)

33 Our Results 7 Multicast Model Eact Tree Algorithm (Distributed) General Graphs O(log n)-approimation NP-Completeness

34 Our Results 7 Multicast Model Eact Tree Algorithm (Distributed) General Graphs O(log n)-approimation NP-Completeness Unicast Model Nonmetric Case O(log n)-approimation Identical Interest Sets / Metric Case O(1)-Approimation NP-Completeness

35 Our Results 7 Multicast Model Eact Tree Algorithm (Distributed) General Graphs O(log n)-approimation NP-Completeness Unicast Model Nonmetric Case O(log n)-approimation Identical Interest Sets / Metric Case O(1)-Approimation NP-Completeness Controlled Broadcast Model A Polnomial LP solution A Combinatorial solution

36 The Multicast Model With Aggregation 8 want the following A push subtree T i for each source i A pull subtree T j for each sink j Whenever j is interested in i (i I j ), T i T j. Total cost of all trees (summing edge weights in each tree) is minimized. For Trees: Basic idea: for each edge, compute minimum possible cost for connectedness of trees. Claim: Global optimum consists of this solution at ever edge.

37 The Multicast Model 9 Indicator variable uvi sas whether uv T i (push tree i) uvj indicates uv T j (pull tree j) z uvij indicates i I j and uv P (T i T j, j) arbitrar m ij is average response frequenc Minimize Objective function p i uv uvi + uv Ec q j uv uvj + j Q uv Ec i P i P j Q m ij c uv z uvij uv E

38 Multicast Model An Eact (Distributed) Tree Algorithm 10 G is a tree T = (V, E) MinC(T i T j, j) is sum of edge weights on shortest path P (T i, j) For edge uv, let S uv be largest subtree containing u but not v Note S vu = V \ S uv Substituting V = S uv S vu, we obtain two smmetric terms (eg.): c uv p i uvi + q j uvj + m ij z uvij i S uv j S vu j S vu uv E i S uv Claim: Global optimum minimizes [...] independentl!

39 Tree Algorithm Diagram 11 a b c z Interest sets: {, z} want {a, b, c}; wants onl a.

40 Tree Algorithm Diagram 11 a b c z Interest sets: {, z} want {a, b, c}; wants onl a.

41 Tree Algorithm Diagram 11 a b c v w z Interest sets: {, z} want {a, b, c}; wants onl a. Question: What is the minimum we can pa on edge vw?

42 Tree Algorithm Diagram 11 a b a b c z c v w z Interest sets: {, z} want {a, b, c}; wants onl a. Question: What is the minimum we can pa on edge vw?

43 Bipartite Minimum Weight Verte Cover 12 a b c z

44 Bipartite Minimum Weight Verte Cover 12 a b c z Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R)

45 Bipartite Minimum Weight Verte Cover 12 a s b z t c Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R)

46 Bipartite Minimum Weight Verte Cover 12 s a b c z t Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R)

47 Bipartite Minimum Weight Verte Cover 12 A B s a b c z t Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R)

48 Bipartite Minimum Weight Verte Cover 12 a b c z Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R) Application: Set A = P vw and B = Q vw

49 Bipartite Minimum Weight Verte Cover 12 a b c z Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R) Application: Set A = P vw and B = Q vw { ij (i, j) X vw } for response costs.

50 Bipartite Minimum Weight Verte Cover 12 a b c z Well Known: For bipartite G vw = (A B, E), MWVC P (Ma flow). Find min cut R, to get MWVC C vw = (A \ R) (B R) Application: Set A = P vw and B = Q vw { ij (i, j) X vw } for response costs. Lemma 1. For each arc e = vw, the MWVC weight of G vw is the minimum value paid for vw in an optimal solution.

51 Tree Algorithm Tiebreaking 13 a b c z Long chain, no sources or sinks.

52 Tree Algorithm Tiebreaking 13 a a b z a b z a b z b c c c c z Long chain, no sources or sinks. Identical Bipartite graph problem

53 Tree Algorithm Tiebreaking 13 a a b z a b z a b z b c c c c z Long chain, no sources or sinks. Identical Bipartite graph problem Suppose man possible MWVCs (eg a + b + c = a + z = + z). How to break MWVC ties?

54 Tree Algorithm Tiebreaking 13 a a b z a b z a b z b c c c c z Long chain, no sources or sinks. Identical Bipartite graph problem Suppose man possible MWVCs (eg a + b + c = a + z = + z). How to break MWVC ties? Defn: In bipartite G = (A B, E), an MWVC is A-maimum if it has maimum weight in A.

55 A Consistent Tiebreaking Solution 14 Defn: In bipartite G = (A B, E), an MWVC is A-maimum if it has maimum weight in A.

56 A Consistent Tiebreaking Solution 14 Defn: In bipartite G = (A B, E), an MWVC is A-maimum if it has maimum weight in A. A E A 1 B 1 B A 2 B 2 Lemma 2. Let G = (A B, E), let A 1, A 2 A and let B 1, B 2 B. If A 1 B 1 and A 2 B 2 are both A-maimum minimum weight verte covers,...

57 A Consistent Tiebreaking Solution 14 Defn: In bipartite G = (A B, E), an MWVC is A-maimum if it has maimum weight in A. A E B A 1 B 1 A 2 B 2 Lemma 2. Let G = (A B, E), let A 1, A 2 A and let B 1, B 2 B. If A 1 B 1 and A 2 B 2 are both A-maimum minimum weight verte covers,... then A 1 = A 2 and B 1 = B 2. Unique solution per edge!

58 Tree Algorithm Structural Continuit 15 a b a b c z c v w z Interest sets recall: {, z} want {a, b, c}; wants onl a.

59 Tree Algorithm Structural Continuit 15 a b u c v w z Interest sets recall: {, z} want {a, b, c}; wants onl a. What about G uv? Clearl different.

60 Tree Algorithm Structural Continuit 15 a a b z b u c v w z Interest sets recall: {, z} want {a, b, c}; wants onl a. What about G uv? Clearl different.

61 Tree Algorithm Structural Continuit 15 a b u a b z a b c z c v w z Interest sets recall: {, z} want {a, b, c}; wants onl a. What about G uv? Clearl different.

62 Tree Algorithm Structural Continuit 15 a b u a b z a b c z c v w z Interest sets recall: {, z} want {a, b, c}; wants onl a. What about G uv? Clearl different. Are push trees, pull trees and response paths connected? Lemma 3. If we compute push-maimum MWVC for ever edge, then Push and Pull subtrees are connected.

63 Yes!!! a b u a b Structural Continuit Solution a z b z c 16 c v w z

64 Yes!!! a b Structural Continuit Solution a z b z c 16

65 Yes!!! a b a b z Structural Continuit Solution a z b z c 16 c

66 Yes!!! Structural Continuit Solution 16 a b c z

67 Yes!!! Structural Continuit Solution A B 16 a b c z Lemma 4. Let uvw be two consecutive edges, let A be the set of push nodes in G uv, and let B be the set of (non-push) nodes in G vw.

68 Yes!!! Structural Continuit Solution A A 1 B 1 B 16 a b c z Lemma 4. Let uvw be two consecutive edges, let A be the set of push nodes in G uv, and let B be the set of (non-push) nodes in G vw. Let A 1, B 1 be parts of push-maimum MWVC of G vw in A, B resp., and

69 Yes!!! Structural Continuit Solution A A B 1 B 1 16 a b z A 2 B 2 c Lemma 4. Let uvw be two consecutive edges, let A be the set of push nodes in G uv, and let B be the set of (non-push) nodes in G vw. Let A 1, B 1 be parts of push-maimum MWVC of G vw in A, B resp., and A 2, B 2 be parts of push-maimum MWVC of G uv in A, B resp.

70 Yes!!! Structural Continuit Solution A A 2 B 1 B 16 a b z A 1 B 2 c Lemma 4. Let uvw be two consecutive edges, let A be the set of push nodes in G uv, and let B be the set of (non-push) nodes in G vw. Let A 1, B 1 be parts of push-maimum MWVC of G vw in A, B resp., and A 2, B 2 be parts of push-maimum MWVC of G uv in A, B resp. then A 1 A 2 and B 1 B 2. Push/Pull subtrees, Response paths are connected!

71 Tree Algorithm 17 for each directed edge uv construct the graph G uv find its canonical minimum cut C uv for all i P uv if i C uv then include uv in T i for all j Q vu if j C uv then include uv in T j for all (i, j) X uv if ij C uv then include uv in P (T i, j)

72 Distributed Implementation 18 Global All-to-all echange of sets of push nodes frequencies, pull nodes frequencies and interest sets. Locall, each edge solves both its directions independentl. Use the solution to push and pull information Notes: Cost of first phase small compared to third. For small sets of distinct values, communication improved.

73 Multicast Model General Graph Approimation algorithm 19 Reduction from Min Steiner Tree; NP-hard to approimate within 96/95. Chlebìk & Chlebìkovà SWAT 02 Theorem 1. There is an epected O(log n)-approimation for the Multicast problem in general graphs.

74 Multicast Model General Graph Approimation algorithm 19 Reduction from Min Steiner Tree; NP-hard to approimate within 96/95. Chlebìk & Chlebìkovà SWAT 02 Theorem 1. There is an epected O(log n)-approimation for the Multicast problem in general graphs. We use the following: Theorem 2 (Fakcharoenphol et al. STOC 03). The distribution over tree metrics resulting from (their) algorithm O(log n)-probabilisticall approimates the metric d.

75 General Graph Approimation algorithm ctd. 20 Bound Derivation Choose T randoml from distribution of metric-spanning trees. Project structures in G into T. Obtain feasible solution for T. OP T (T ) O(log n) OP T (G). Approimation Algorithm Solve T eactl using our algorithm. Project structures in T into G. Obtain feasible solution for G. ALG(G) 2 OP T (T )

76 General Graph Approimation algorithm ctd. 20 Bound Derivation Choose T randoml from distribution of metric-spanning trees. Project structures in G into T. Obtain feasible solution for T. OP T (T ) O(log n) OP T (G). Approimation Algorithm Solve T eactl using our algorithm. Project structures in T into G. Obtain feasible solution for G. ALG(G) 2 OP T (T ) O(log n) OP T (G).

77 Multicast Model Hardness 21 Multicast problem with(out) aggregation: eas reduction from Min Steiner tree. Arbitrar node becomes low-freq source Rest become high-freq Sink nodes Each interested in Source

78 Multicast Model Hardness 21 Multicast problem with(out) aggregation: eas reduction from Min Steiner tree. Arbitrar node becomes low-freq source Rest become high-freq Sink nodes Each interested in Source

79 Multicast Model Hardness 21 Multicast problem with(out) aggregation: eas reduction from Min Steiner tree. Arbitrar node becomes low-freq source Rest become high-freq Sink nodes Each interested in Source Min Steiner Tree NP-hard to approimate within 96/95. Chlebìk & Chlebìkovà SWAT 02

80 The Unicast Model 22

81 The Unicast Model 22 Given (non-)metric distances d uv for ever pair (u, v) V V. SetC(u, S) = k S d uk find push-sets P i and pull-sets Q j that minimize total communication cost: p i d ik + q j d kj + q j RespC(j), i P k P i j Q k Q j j Q and satisfies: for all i I j, P i Q j where RespC(j) = { SetC(j, Qj ) (aggregation model) i I j MinC(P i Q j, j) otherwise.

82 Unicast Model with Aggregation An Integer Program 23 Replace response cost b doubling sink frequencies ik = 1 means i pushes to k kj = 1 means j pulls from k r ijk = 1 means i talks to j through k. Minimize: p i d ik ik + q j d kj kj i P j Q k V k V subject to r ijk ik r ijk kj k r ijk 1, where ik, kj, r ijk {0, 1}.

83 Unicast Model with Aggregation Nonmetric Case via Randomized Rounding 24 Convert to LP: Use 0 instead of {0, 1} Solve and discard values 1/n 2 and scale b n/(n 1) Round values up to powers of 1/2, obtain (, ỹ, z) For node k and 0 p < 2 log n, define X pk as i such that ik 1/2 p. p, k: with probabilit min{1, (log n)/2 p } add k to P i and Q j for all i X pk and j Y pk. Theorem 3. With high probabilit, solution is feasible, with cost O(log n) OP T LP.

84 Unicast Model with Aggregation Nonmetric Case via Randomized Rounding Proof 25 Since i X log( rijk )k and j Y log( rijk )k, Pr[k P i Q j ] min{1, r ijk log(n)}. Clearl Pr[k P i ] p:i X pk (log n)/2 p = 2 ik log n min{1, 2 ik log n}. Pr[P i Q j = ] = k (1 r ijk log n) e P k r ijk log n 1/n 2. Define r.v. C i as push cost for i, and r.v. C ik takes value d ik with probabilit min{1, 2 ik log n}. Chernoff-Hoeffding: w.h.p. k C ik O(log n) k d ik ik. Summing over all sources, sinks gives cost bound w.h.p.

85 Unicast Model with Aggregation Uniform Interests, Metric Case O(1)-Approimation 26 Overview Applies for Identical/Disjoint Interest Sets Uses same Integer Program. Deterministic Rounding with Filtering Technique Lin & Vitter IPL 92, Shmos et al STOC 97, Ravi & Sinha SODA 04

86 Unicast Model with Aggregation Uniform Interest Sets in Metric Case Intro 27 Basic definitions Optimal solution to the LP is (,, r ). LP gives cost lower bounds C i = k d ik ik and C j = k d kj kj 2*(1/2) C = 6.5 j j 3*(1/2) 4*(1)

87 Unicast Model with Aggregation Uniform Interest Sets in Metric Case Intro 27 Basic definitions Optimal solution to the LP is (,, r ). LP gives cost lower bounds C i = k d ik ik and C j = k d kj kj For node u, r > 0, define B u (r) = {v : d uv r}. Let 1 < α < β. Clearl B j (C j ) B j(αc j ) B j(βc j ) j

88 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

89 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

90 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

91 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

92 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

93 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

94 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

95 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

96 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost.

97 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost. Define push sets: P i = {i} {l i } {j : j S and C j C i} and pull sets: Q j = {j} {l j } {i : i S and C i < C j }. j

98 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm 28 Choose leaders: nodes with disjoint β-balls, b nondecreasing cost. Define push sets: P i = {i} {l i } {j : j S and C j C i} and pull sets: Q j = {j} {l j } {i : i S and C i < C j }. Intersection guarantee: For each i P and j Q, P i Q j. m i j

99 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm Proof Relative distance limits total push etent: For i P, α > 1, ik 1/α k / B i (αc i ) 29 j C j 4C j

100 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm Proof Relative distance limits total push etent: For i P, α > 1, ik 1/α k / B i (αc i ) Derive Approimation Ratio. Recall: P i = {i} {l i } {j : j S and C j C i} Cost to i s leader l i : 2βC i 29 j l j

101 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm Proof Relative distance limits total push etent: For i P, α > 1, ik 1/α k / B i (αc i ) Derive Approimation Ratio. Recall: P i = {i} {l i } {j : j S and C j C i} Cost to i s leader l i : 2βC i Cost to (other) leaders S i : 29 C i j S i (d ij αc j ) k B j (αc j ) r ijk i j

102 Unicast Model with Aggregation Uniform Interest Set / Metric Algorithm Proof Relative distance limits total push etent: For i P, α > 1, ik 1/α k / B i (αc i ) Derive Approimation Ratio. Recall: P i = {i} {l i } {j : j S and C j C i} Cost to i s leader l i : 2βC i Cost to (other) leaders S i : 29 C i j S i (d ij αc j ) k B j (αc j ) r ijk [1 ] j S i d ij [1 β] α 1 α j S i d ij. = (β α)(α 1) αβ α = 1.69 and β = 2.86 obtains approimation.

103 Conclusions and Open Problems 30 Nonuniform Packet Lengths Multicast: General Graphs; Can O(log n) UB be improved to O(1)? Nonmetric Unicast: Derandomizing O(log n) algorithm. Close gap O(1) LB vs O(log n) UB gap Metric Unicast Case Improving the bound for Uniform Interest sets. Non-uniform interest sets (UB and/or Hardness) Dnamic Graphs Frequenc, Position and Topolog changes

104 Thank You! 31 Chakinala, Kumarasubramanian, and Manokaran: partial support generous gift from Northeastern Universit alumnus Madhav Anand. Rajaraman, partial support NSF grant IIS Laing, partial support the Mellon Foundation and the Facult Research Awards Committee of Tufts Universit, while visiting Northeastern Universit.

The Steiner Ratio for Obstacle-Avoiding Rectilinear Steiner Trees

The Steiner Ratio for Obstacle-Avoiding Rectilinear Steiner Trees The Steiner Ratio for Obstacle-Avoiding Rectilinear Steiner Trees Anna Lubiw Mina Razaghpour Abstract We consider the problem of finding a shortest rectilinear Steiner tree for a given set of pointn the

More information

On Fixed Cost k-flow Problems

On Fixed Cost k-flow Problems On Fixed Cost k-flow Problems MohammadTaghi Hajiaghayi 1, Rohit Khandekar 2, Guy Kortsarz 3, and Zeev Nutov 4 1 University of Maryland, College Park, MD. hajiagha@cs.umd.edu. 2 Knight Capital Group, Jersey

More information

arxiv: v1 [cs.ds] 20 Nov 2016

arxiv: v1 [cs.ds] 20 Nov 2016 Algorithmic and Hardness Results for the Hub Labeling Problem Haris Angelidakis 1, Yury Makarychev 1 and Vsevolod Oparin 2 1 Toyota Technological Institute at Chicago 2 Saint Petersburg Academic University

More information

Designing Overlapping Networks for Publish-Subscribe Systems

Designing Overlapping Networks for Publish-Subscribe Systems Designing Overlapping Networks for Publish-Subscribe Systems Jennifer Iglesias 1, Rajmohan Rajaraman 2, R. Ravi 1, and Ravi Sundaram 2 1 Carnegie Mellon University 5000 Forbes Avenue, Pittsburgh, PA, USA

More information

Toda s Theorem: PH P #P

Toda s Theorem: PH P #P CS254: Computational Compleit Theor Prof. Luca Trevisan Final Project Ananth Raghunathan 1 Introduction Toda s Theorem: PH P #P The class NP captures the difficult of finding certificates. However, in

More information

Steiner Forest Orientation Problems

Steiner Forest Orientation Problems Steiner Forest Orientation Problems Marek Cygan Guy Kortsarz Zeev Nutov November 23, 2011 Abstract We consider connectivity problems with orientation constraints. Given a directed graph D and a collection

More information

On the Fixed Cost k-flow Problem and related problems

On the Fixed Cost k-flow Problem and related problems On the Fixed Cost k-flow Problem and related problems MohammadTaghi Hajiaghayi 1, Rohit Khandekar 2, Guy Kortsarz 3, and Zeev Nutov 4 1 University of Maryland, College Park, MD. hajiagha@cs.umd.edu. 2

More information

CMPUT 675: Approximation Algorithms Fall 2014

CMPUT 675: Approximation Algorithms Fall 2014 CMPUT 675: Approximation Algorithms Fall 204 Lecture 25 (Nov 3 & 5): Group Steiner Tree Lecturer: Zachary Friggstad Scribe: Zachary Friggstad 25. Group Steiner Tree In this problem, we are given a graph

More information

- Well-characterized problems, min-max relations, approximate certificates. - LP problems in the standard form, primal and dual linear programs

- Well-characterized problems, min-max relations, approximate certificates. - LP problems in the standard form, primal and dual linear programs LP-Duality ( Approximation Algorithms by V. Vazirani, Chapter 12) - Well-characterized problems, min-max relations, approximate certificates - LP problems in the standard form, primal and dual linear programs

More information

Steiner Forest Orientation Problems

Steiner Forest Orientation Problems Steiner Forest Orientation Problems Marek Cygan Guy Kortsarz Zeev Nutov November 20, 2011 Abstract We consider connectivity problems with orientation constraints. Given a directed graph D and a collection

More information

Capacitated Network Design Problems: Hardness, Approximation Algorithms, and Connections to Group Steiner Tree

Capacitated Network Design Problems: Hardness, Approximation Algorithms, and Connections to Group Steiner Tree Capacitated Network Design Problems: Hardness, Approximation Algorithms, and Connections to Group Steiner Tree MohammadTaghi Hajiaghayi 1, Rohit Khandekar 2, GuyKortsarz 3, and Zeev Nutov 4 1 University

More information

Metric Approximations (Embeddings) M M 1 2 f M 1, M 2 metric spaces. f injective mapping from M 1 to M 2. M 2 dominates M 1 under f if d M1 (u; v) d M

Metric Approximations (Embeddings) M M 1 2 f M 1, M 2 metric spaces. f injective mapping from M 1 to M 2. M 2 dominates M 1 under f if d M1 (u; v) d M Approximating a nite metric by a small number of tree metrics Moses Charikar Chandra Chekuri Ashish Goel Sudipto Guha Serge Plotkin Metric Approximations (Embeddings) M M 1 2 f M 1, M 2 metric spaces.

More information

An Improved Approximation Algorithm for Requirement Cut

An Improved Approximation Algorithm for Requirement Cut An Improved Approximation Algorithm for Requirement Cut Anupam Gupta Viswanath Nagarajan R. Ravi Abstract This note presents improved approximation guarantees for the requirement cut problem: given an

More information

New Hardness Results for Undirected Edge Disjoint Paths

New Hardness Results for Undirected Edge Disjoint Paths New Hardness Results for Undirected Edge Disjoint Paths Julia Chuzhoy Sanjeev Khanna July 2005 Abstract In the edge-disjoint paths (EDP) problem, we are given a graph G and a set of source-sink pairs in

More information

Lecture 7: Advanced Coupling & Mixing Time via Eigenvalues

Lecture 7: Advanced Coupling & Mixing Time via Eigenvalues Counting and Sampling Fall 07 Lecture 7: Advanced Coupling & Miing Time via Eigenvalues Lecturer: Shaan Oveis Gharan October 8 Disclaimer: These notes have not been subjected to the usual scrutin reserved

More information

Iterative rounding approximation algorithms for degree-bounded node-connectivity network design

Iterative rounding approximation algorithms for degree-bounded node-connectivity network design Iterative rounding approximation algorithms for degree-bounded node-connectivity network design Takuro Fukunaga Graduate School of Informatics, Kyoto University, Japan Email: takuro@amp.i.kyoto-u.ac.jp

More information

Approximating Minimum-Power Degree and Connectivity Problems

Approximating Minimum-Power Degree and Connectivity Problems Approximating Minimum-Power Degree and Connectivity Problems Guy Kortsarz Vahab S. Mirrokni Zeev Nutov Elena Tsanko Abstract Power optimization is a central issue in wireless network design. Given a graph

More information

Transactions on Algorithms. Two-stage Robust Network Design with Exponential Scenarios

Transactions on Algorithms. Two-stage Robust Network Design with Exponential Scenarios Transactions on Algorithms Two-stage Robust Networ Design with Exponential Scenarios Journal: Transactions on Algorithms Manuscript ID: draft Manuscript Type: Original Article Date Submitted by the Author:

More information

Maximum-Life Routing Schedule

Maximum-Life Routing Schedule Maximum-Life Routing Schedule Peng-Jun Wan wan@cs.iit.edu Peng-Jun Wan (wan@cs.iit.edu) Maximum-Life Routing Schedule 1 / 42 Outline Problem Description Min-Cost Routing Ellipsoid Algorithm Price-Directive

More information

Chapter 6. Self-Adjusting Data Structures

Chapter 6. Self-Adjusting Data Structures Chapter 6 Self-Adjusting Data Structures Chapter 5 describes a data structure that is able to achieve an epected quer time that is proportional to the entrop of the quer distribution. The most interesting

More information

The Steiner k-cut Problem

The Steiner k-cut Problem The Steiner k-cut Problem Chandra Chekuri Sudipto Guha Joseph (Seffi) Naor September 23, 2005 Abstract We consider the Steiner k-cut problem which generalizes both the k-cut problem and the multiway cut

More information

Robust Network Codes for Unicast Connections: A Case Study

Robust Network Codes for Unicast Connections: A Case Study Robust Network Codes for Unicast Connections: A Case Study Salim Y. El Rouayheb, Alex Sprintson, and Costas Georghiades Department of Electrical and Computer Engineering Texas A&M University College Station,

More information

EECS 495: Randomized Algorithms Lecture 14 Random Walks. j p ij = 1. Pr[X t+1 = j X 0 = i 0,..., X t 1 = i t 1, X t = i] = Pr[X t+

EECS 495: Randomized Algorithms Lecture 14 Random Walks. j p ij = 1. Pr[X t+1 = j X 0 = i 0,..., X t 1 = i t 1, X t = i] = Pr[X t+ EECS 495: Randomized Algorithms Lecture 14 Random Walks Reading: Motwani-Raghavan Chapter 6 Powerful tool for sampling complicated distributions since use only local moves Given: to explore state space.

More information

Separating Hierarchical and General Hub Labelings

Separating Hierarchical and General Hub Labelings Separating Hierarchical and General Hub Labelings Andrew V. Goldberg 1 Ilya Razenshteyn 2 Ruslan Savchenko 3 1 Microsoft Research Silicon Valley 2 CSAIL, MIT 3 Moscow State University June 23, 2013 Overview

More information

Design and Analysis of Algorithms

Design and Analysis of Algorithms CSE 0, Winter 208 Design and Analsis of Algorithms Lecture 6: NP-Completeness, Part 2 Class URL: http://vlsicad.ucsd.edu/courses/cse0-w8/ The Classes P and NP Classes of Decision Problems P: Problems for

More information

Multi-criteria approximation schemes for the resource constrained shortest path problem

Multi-criteria approximation schemes for the resource constrained shortest path problem Noname manuscript No. (will be inserted by the editor) Multi-criteria approximation schemes for the resource constrained shortest path problem Markó Horváth Tamás Kis Received: date / Accepted: date Abstract

More information

Topic: Balanced Cut, Sparsest Cut, and Metric Embeddings Date: 3/21/2007

Topic: Balanced Cut, Sparsest Cut, and Metric Embeddings Date: 3/21/2007 CS880: Approximations Algorithms Scribe: Tom Watson Lecturer: Shuchi Chawla Topic: Balanced Cut, Sparsest Cut, and Metric Embeddings Date: 3/21/2007 In the last lecture, we described an O(log k log D)-approximation

More information

Steiner Transitive-Closure Spanners of Low-Dimensional Posets

Steiner Transitive-Closure Spanners of Low-Dimensional Posets Steiner Transitive-Closure Spanners of Low-Dimensional Posets Grigory Yaroslavtsev Pennsylvania State University + AT&T Labs Research (intern) Joint with P. Berman (PSU), A. Bhattacharyya (MIT), E. Grigorescu

More information

PCPs and Inapproximability Gap-producing and Gap-Preserving Reductions. My T. Thai

PCPs and Inapproximability Gap-producing and Gap-Preserving Reductions. My T. Thai PCPs and Inapproximability Gap-producing and Gap-Preserving Reductions My T. Thai 1 1 Hardness of Approximation Consider a maximization problem Π such as MAX-E3SAT. To show that it is NP-hard to approximation

More information

On the Crossing Spanning Tree Problem

On the Crossing Spanning Tree Problem On the Crossing Spanning Tree Problem Vittorio Bilò 1, Vineet Goyal,RRavi, and Mohit Singh 1 Dipartimento di Informatica Università di L Aquila. Via Vetoio, Coppito 67100 L Aquila. Italy. bilo@di.univaq.it

More information

Approximability of Packing Disjoint Cycles

Approximability of Packing Disjoint Cycles Approximability of Packing Disjoint Cycles Zachary Friggstad Mohammad R. Salavatipour Department of Computing Science University of Alberta Edmonton, Alberta T6G 2E8, Canada zacharyf,mreza@cs.ualberta.ca

More information

Approximating Graph Spanners

Approximating Graph Spanners Approximating Graph Spanners Michael Dinitz Johns Hopkins University Joint work with combinations of Robert Krauthgamer (Weizmann), Eden Chlamtáč (Ben Gurion), Ran Raz (Weizmann), Guy Kortsarz (Rutgers-Camden)

More information

Speeding up the Dreyfus-Wagner Algorithm for minimum Steiner trees

Speeding up the Dreyfus-Wagner Algorithm for minimum Steiner trees Speeding up the Dreyfus-Wagner Algorithm for minimum Steiner trees Bernhard Fuchs Center for Applied Computer Science Cologne (ZAIK) University of Cologne, Weyertal 80, 50931 Köln, Germany Walter Kern

More information

Network Design and Game Theory Spring 2008 Lecture 6

Network Design and Game Theory Spring 2008 Lecture 6 Network Design and Game Theory Spring 2008 Lecture 6 Guest Lecturer: Aaron Archer Instructor: Mohammad T. Hajiaghayi Scribe: Fengming Wang March 3, 2008 1 Overview We study the Primal-dual, Lagrangian

More information

16.1 Min-Cut as an LP

16.1 Min-Cut as an LP 600.469 / 600.669 Approximation Algorithms Lecturer: Michael Dinitz Topic: LPs as Metrics: Min Cut and Multiway Cut Date: 4//5 Scribe: Gabriel Kaptchuk 6. Min-Cut as an LP We recall the basic definition

More information

Hardness of Embedding Metric Spaces of Equal Size

Hardness of Embedding Metric Spaces of Equal Size Hardness of Embedding Metric Spaces of Equal Size Subhash Khot and Rishi Saket Georgia Institute of Technology {khot,saket}@cc.gatech.edu Abstract. We study the problem embedding an n-point metric space

More information

arxiv:cs/ v1 [cs.cc] 7 Sep 2006

arxiv:cs/ v1 [cs.cc] 7 Sep 2006 Approximation Algorithms for the Bipartite Multi-cut problem arxiv:cs/0609031v1 [cs.cc] 7 Sep 2006 Sreyash Kenkre Sundar Vishwanathan Department Of Computer Science & Engineering, IIT Bombay, Powai-00076,

More information

Distributed primal-dual approximation algorithms for network design problems

Distributed primal-dual approximation algorithms for network design problems Distributed primal-dual approximation algorithms for network design problems Zeev Nutov The Open University of Israel nutov@openu.ac.il Amir Sadeh The Open University of Israel amirsadeh@yahoo.com Abstract

More information

Distributed Distance-Bounded Network Design Through Distributed Convex Programming

Distributed Distance-Bounded Network Design Through Distributed Convex Programming Distributed Distance-Bounded Network Design Through Distributed Convex Programming OPODIS 2017 Michael Dinitz, Yasamin Nazari Johns Hopkins University December 18, 2017 Distance Bounded Network Design

More information

CSE 5311 Notes 5: Trees. (Last updated 6/4/13 4:12 PM)

CSE 5311 Notes 5: Trees. (Last updated 6/4/13 4:12 PM) SE 511 Notes 5: Trees (Last updated 6//1 :1 PM) What is the optimal wa to organie a static tree for searching? n optimal (static) binar search tree is significantl more complicated to construct than an

More information

CS 6820 Fall 2014 Lectures, October 3-20, 2014

CS 6820 Fall 2014 Lectures, October 3-20, 2014 Analysis of Algorithms Linear Programming Notes CS 6820 Fall 2014 Lectures, October 3-20, 2014 1 Linear programming The linear programming (LP) problem is the following optimization problem. We are given

More information

3 Polynomial and Rational Functions

3 Polynomial and Rational Functions 3 Polnomial and Rational Functions 3.1 Quadratic Functions and Models 3.2 Polnomial Functions and Their Graphs 3.3 Dividing Polnomials 3.4 Real Zeros of Polnomials 3.5 Comple Zeros and the Fundamental

More information

Partitioning Metric Spaces

Partitioning Metric Spaces Partitioning Metric Spaces Computational and Metric Geometry Instructor: Yury Makarychev 1 Multiway Cut Problem 1.1 Preliminaries Definition 1.1. We are given a graph G = (V, E) and a set of terminals

More information

Near-Optimal Hardness Results and Approximation Algorithms for Edge-Disjoint Paths and Related Problems

Near-Optimal Hardness Results and Approximation Algorithms for Edge-Disjoint Paths and Related Problems Near-Optimal Hardness Results and Approximation Algorithms for Edge-Disjoint Paths and Related Problems Venkatesan Guruswami Sanjeev Khanna Rajmohan Rajaraman Bruce Shepherd Mihalis Yannakakis Abstract

More information

Scheduling and Optimization Course (MPRI)

Scheduling and Optimization Course (MPRI) MPRI Scheduling and optimization: lecture p. /6 Scheduling and Optimization Course (MPRI) Leo Liberti LIX, École Polytechnique, France MPRI Scheduling and optimization: lecture p. /6 Teachers Christoph

More information

ACO Comprehensive Exam 19 March Graph Theory

ACO Comprehensive Exam 19 March Graph Theory 1. Graph Theory Let G be a connected simple graph that is not a cycle and is not complete. Prove that there exist distinct non-adjacent vertices u, v V (G) such that the graph obtained from G by deleting

More information

Lecturer: Shuchi Chawla Topic: Inapproximability Date: 4/27/2007

Lecturer: Shuchi Chawla Topic: Inapproximability Date: 4/27/2007 CS880: Approximations Algorithms Scribe: Tom Watson Lecturer: Shuchi Chawla Topic: Inapproximability Date: 4/27/2007 So far in this course, we have been proving upper bounds on the approximation factors

More information

A PARTIAL CHARACTERIZATION OF THE COCIRCUITS OF A SPLITTING MATROID

A PARTIAL CHARACTERIZATION OF THE COCIRCUITS OF A SPLITTING MATROID DEPARTMENT OF MATHEMATICS TECHNICAL REPORT A PARTIAL CHARACTERIZATION OF THE COCIRCUITS OF A SPLITTING MATROID DR. ALLAN D. MILLS MAY 2004 No. 2004-2 TENNESSEE TECHNOLOGICAL UNIVERSITY Cookeville, TN 38505

More information

The 2-valued case of makespan minimization with assignment constraints

The 2-valued case of makespan minimization with assignment constraints The 2-valued case of maespan minimization with assignment constraints Stavros G. Kolliopoulos Yannis Moysoglou Abstract We consider the following special case of minimizing maespan. A set of jobs J and

More information

A 1.8 approximation algorithm for augmenting edge-connectivity of a graph from 1 to 2

A 1.8 approximation algorithm for augmenting edge-connectivity of a graph from 1 to 2 A 1.8 approximation algorithm for augmenting edge-connectivity of a graph from 1 to 2 Guy Even Jon Feldman Guy Kortsarz Zeev Nutov October 20, 2008 Abstract We present a 1.8-approximation algorithm for

More information

Constant 2-labelling of a graph

Constant 2-labelling of a graph Constant 2-labelling of a graph S. Gravier, and E. Vandomme June 18, 2012 Abstract We introduce the concept of constant 2-labelling of a graph and show how it can be used to obtain periodic sphere packing.

More information

Chapter 2 Polynomial, Power, and Rational Functions

Chapter 2 Polynomial, Power, and Rational Functions Section. Linear and Quadratic Functions and Modeling 6 Chapter Polnomial, Power, and Rational Functions Section. Linear and Quadratic Functions and Modeling Eploration. $000 per ear.. The equation will

More information

An 1.75 approximation algorithm for the leaf-to-leaf tree augmentation problem

An 1.75 approximation algorithm for the leaf-to-leaf tree augmentation problem An 1.75 approximation algorithm for the leaf-to-leaf tree augmentation problem Zeev Nutov, László A. Végh January 12, 2016 We study the tree augmentation problem: Tree Augmentation Problem (TAP) Instance:

More information

Classical Complexity and Fixed-Parameter Tractability of Simultaneous Consecutive Ones Submatrix & Editing Problems

Classical Complexity and Fixed-Parameter Tractability of Simultaneous Consecutive Ones Submatrix & Editing Problems Classical Complexity and Fixed-Parameter Tractability of Simultaneous Consecutive Ones Submatrix & Editing Problems Rani M. R, Mohith Jagalmohanan, R. Subashini Binary matrices having simultaneous consecutive

More information

Design and Analysis of Algorithms

Design and Analysis of Algorithms CSE 101, Winter 2018 Design and Analsis of Algorithms Lecture 15: NP-Completeness Class URL: http://vlsicad.ucsd.edu/courses/cse101-w18/ From Application to Algorithm Design Find all maimal regularl-spaced,

More information

TRANSPORTATION PROBLEMS

TRANSPORTATION PROBLEMS Chapter 6 TRANSPORTATION PROBLEMS 61 Transportation Model Transportation models deal with the determination of a minimum-cost plan for transporting a commodity from a number of sources to a number of destinations

More information

The Generalized Regenerator Location Problem

The Generalized Regenerator Location Problem The Generalized Regenerator Location Problem Si Chen Ivana Ljubić S. Raghavan College of Business and Public Affairs, Murray State University Murray, KY 42071, USA Faculty of Business, Economics, and Statistics,

More information

The Polymatroid Steiner Problems

The Polymatroid Steiner Problems The Polymatroid Steiner Problems G. Calinescu 1 and A. Zelikovsky 2 1 Department of Computer Science, Illinois Institute of Technology, Chicago, IL 60616. E-mail: calinesc@iit.edu. 2 Department of Computer

More information

Disjoint Bases in a Polymatroid

Disjoint Bases in a Polymatroid Disjoint Bases in a Polymatroid Gruia Călinescu Chandra Chekuri Jan Vondrák May 26, 2008 Abstract Let f : 2 N Z + be a polymatroid (an integer-valued non-decreasing submodular set function with f( ) =

More information

New Approaches to Multi-Objective Optimization

New Approaches to Multi-Objective Optimization New Approaches to Multi-Objective Optimization Fabrizio Grandoni R. Ravi Mohit Singh Rico Zenklusen July 17, 2013 Abstract A natural way to deal with multiple, partially conflicting objectives is turning

More information

Lecture 14: SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices) Lecturer: Sanjeev Arora

Lecture 14: SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices) Lecturer: Sanjeev Arora princeton univ. F 13 cos 521: Advanced Algorithm Design Lecture 14: SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices) Lecturer: Sanjeev Arora Scribe: Today we continue the

More information

Approximation Hardness for Small Occurrence Instances of NP-Hard Problems

Approximation Hardness for Small Occurrence Instances of NP-Hard Problems Approximation Hardness for Small Occurrence Instances of NP-Hard Problems Miroslav Chlebík and Janka Chlebíková 2 MPI for Mathematics in the Sciences, D-0403 Leipzig 2 CAU, Institut für Informatik und

More information

Transportation Problem

Transportation Problem Transportation Problem Alireza Ghaffari-Hadigheh Azarbaijan Shahid Madani University (ASMU) hadigheha@azaruniv.edu Spring 2017 Alireza Ghaffari-Hadigheh (ASMU) Transportation Problem Spring 2017 1 / 34

More information

Polyhedral Approaches to Online Bipartite Matching

Polyhedral Approaches to Online Bipartite Matching Polyhedral Approaches to Online Bipartite Matching Alejandro Toriello joint with Alfredo Torrico, Shabbir Ahmed Stewart School of Industrial and Systems Engineering Georgia Institute of Technology Industrial

More information

Prize-collecting Survivable Network Design in Node-weighted Graphs. Chandra Chekuri Alina Ene Ali Vakilian

Prize-collecting Survivable Network Design in Node-weighted Graphs. Chandra Chekuri Alina Ene Ali Vakilian Prize-collecting Survivable Network Design in Node-weighted Graphs Chandra Chekuri Alina Ene Ali Vakilian Survivable Network Design (SNDP) Collection of l pairs: s 1, t 1,, (s l, t l ) r(s i, t i ): requirement

More information

On the Construction of Data Aggregation Tree with Minimum Energy Cost in Wireless Sensor Networks: NP-Completeness and Approximation Algorithms

On the Construction of Data Aggregation Tree with Minimum Energy Cost in Wireless Sensor Networks: NP-Completeness and Approximation Algorithms On the Construction of Data Aggregation Tree with Minimum Energy Cost in Wireless Sensor Networks: NP-Completeness and Approximation Algorithms Tung-Wei Kuo and Ming-Jer Tsai Department of Computer Science

More information

Blockers for the stability number and the chromatic number

Blockers for the stability number and the chromatic number Blockers for the stability number and the chromatic number C. Bazgan C. Bentz C. Picouleau B. Ries Abstract Given an undirected graph G = (V, E) and two positive integers k and d, we are interested in

More information

Extreme Point Solutions for Infinite Network Flow Problems

Extreme Point Solutions for Infinite Network Flow Problems Extreme Point Solutions for Infinite Network Flow Problems H. Edwin Romeijn Dushyant Sharma Robert L. Smith January 3, 004 Abstract We study capacitated network flow problems with supplies and demands

More information

Improved Approximating Algorithms for Directed Steiner Forest

Improved Approximating Algorithms for Directed Steiner Forest Improved Approximating Algorithms for Directed Steiner Forest Moran Feldman Guy Kortsarz Zeev Nutov Abstract We consider the k-directed Steiner Forest (k-dsf) problem: Given a directed graph G = (V, E)

More information

Second main application of Chernoff: analysis of load balancing. Already saw balls in bins example. oblivious algorithms only consider self packet.

Second main application of Chernoff: analysis of load balancing. Already saw balls in bins example. oblivious algorithms only consider self packet. Routing Second main application of Chernoff: analysis of load balancing. Already saw balls in bins example synchronous message passing bidirectional links, one message per step queues on links permutation

More information

Closest String and Closest Substring Problems

Closest String and Closest Substring Problems January 8, 2010 Problem Formulation Problem Statement I Closest String Given a set S = {s 1, s 2,, s n } of strings each length m, find a center string s of length m minimizing d such that for every string

More information

CS5314 Randomized Algorithms. Lecture 15: Balls, Bins, Random Graphs (Hashing)

CS5314 Randomized Algorithms. Lecture 15: Balls, Bins, Random Graphs (Hashing) CS5314 Randomized Algorithms Lecture 15: Balls, Bins, Random Graphs (Hashing) 1 Objectives Study various hashing schemes Apply balls-and-bins model to analyze their performances 2 Chain Hashing Suppose

More information

Proof of Theorem 1. Tao Lei CSAIL,MIT. Here we give the proofs of Theorem 1 and other necessary lemmas or corollaries.

Proof of Theorem 1. Tao Lei CSAIL,MIT. Here we give the proofs of Theorem 1 and other necessary lemmas or corollaries. Proof of Theorem 1 Tao Lei CSAIL,MIT Here we give the proofs of Theorem 1 and other necessary lemmas or corollaries. Lemma 1 (Reachability) Any two trees y, y are reachable to each other. Specifically,

More information

On the Complexity of Budgeted Maximum Path Coverage on Trees

On the Complexity of Budgeted Maximum Path Coverage on Trees On the Complexity of Budgeted Maximum Path Coverage on Trees H.-C. Wirth An instance of the budgeted maximum coverage problem is given by a set of weighted ground elements and a cost weighted family of

More information

Provable Approximation via Linear Programming

Provable Approximation via Linear Programming Chapter 7 Provable Approximation via Linear Programming One of the running themes in this course is the notion of approximate solutions. Of course, this notion is tossed around a lot in applied work: whenever

More information

Notes on MapReduce Algorithms

Notes on MapReduce Algorithms Notes on MapReduce Algorithms Barna Saha 1 Finding Minimum Spanning Tree of a Dense Graph in MapReduce We are given a graph G = (V, E) on V = N vertices and E = m N 1+c edges for some constant c > 0. Our

More information

Duality of LPs and Applications

Duality of LPs and Applications Lecture 6 Duality of LPs and Applications Last lecture we introduced duality of linear programs. We saw how to form duals, and proved both the weak and strong duality theorems. In this lecture we will

More information

a. plotting points in Cartesian coordinates (Grade 9 and 10), b. using a graphing calculator such as the TI-83 Graphing Calculator,

a. plotting points in Cartesian coordinates (Grade 9 and 10), b. using a graphing calculator such as the TI-83 Graphing Calculator, GRADE PRE-CALCULUS UNIT C: QUADRATIC FUNCTIONS CLASS NOTES FRAME. After linear functions, = m + b, and their graph the Quadratic Functions are the net most important equation or function. The Quadratic

More information

Geometric Steiner Trees

Geometric Steiner Trees Geometric Steiner Trees From the book: Optimal Interconnection Trees in the Plane By Marcus Brazil and Martin Zachariasen Part 3: Computational Complexity and the Steiner Tree Problem Marcus Brazil 2015

More information

Branch-and-Bound for the Travelling Salesman Problem

Branch-and-Bound for the Travelling Salesman Problem Branch-and-Bound for the Travelling Salesman Problem Leo Liberti LIX, École Polytechnique, F-91128 Palaiseau, France Email:liberti@lix.polytechnique.fr March 15, 2011 Contents 1 The setting 1 1.1 Graphs...............................................

More information

The Parking Permit Problem

The Parking Permit Problem The Parking Permit Problem Adam Meyerson November 3, 2004 Abstract We consider online problems where purchases have time durations which expire regardless of whether the purchase is used or not. The Parking

More information

Graph independent field size bounds on failure protecting network codes

Graph independent field size bounds on failure protecting network codes Egerváry Research Group on Combinatorial Optimization Technical reports TR-25-. Published by the Egerváry Research Group, Pázmány P. sétány /C, H 7, Budapest, Hungary. Web site: www.cs.elte.hu/egres. ISSN

More information

Collective Tree Spanners of Unit Disk Graphs with Applications to Compact and Low Delay Routing Labeling Schemes. F.F. Dragan

Collective Tree Spanners of Unit Disk Graphs with Applications to Compact and Low Delay Routing Labeling Schemes. F.F. Dragan Collective Tree Spanners of Unit Disk Graphs with Applications to Compact and Low Delay Routing Labeling Schemes F.F. Dragan (joint work with Yang Xiang and Chenyu Yan) Kent State University, USA Well-known

More information

Nova Scotia Examinations Mathematics 12 Web Sample 2. Student Booklet

Nova Scotia Examinations Mathematics 12 Web Sample 2. Student Booklet Nova Scotia Eaminations Mathematics Web Sample Student Booklet General Instructions - WEB SAMPLE* This eamination is composed of two sections with the following suggested time allotment: Selected-Response

More information

Computing and Communicating Functions over Sensor Networks

Computing and Communicating Functions over Sensor Networks Computing and Communicating Functions over Sensor Networks Solmaz Torabi Dept. of Electrical and Computer Engineering Drexel University solmaz.t@drexel.edu Advisor: Dr. John M. Walsh 1/35 1 Refrences [1]

More information

Approximation Algorithms for Asymmetric TSP by Decomposing Directed Regular Multigraphs

Approximation Algorithms for Asymmetric TSP by Decomposing Directed Regular Multigraphs Approximation Algorithms for Asymmetric TSP by Decomposing Directed Regular Multigraphs Haim Kaplan Tel-Aviv University, Israel haimk@post.tau.ac.il Nira Shafrir Tel-Aviv University, Israel shafrirn@post.tau.ac.il

More information

4 Inverse function theorem

4 Inverse function theorem Tel Aviv Universit, 2013/14 Analsis-III,IV 53 4 Inverse function theorem 4a What is the problem................ 53 4b Simple observations before the theorem..... 54 4c The theorem.....................

More information

Lecture 3. 1 Polynomial-time algorithms for the maximum flow problem

Lecture 3. 1 Polynomial-time algorithms for the maximum flow problem ORIE 633 Network Flows August 30, 2007 Lecturer: David P. Williamson Lecture 3 Scribe: Gema Plaza-Martínez 1 Polynomial-time algorithms for the maximum flow problem 1.1 Introduction Let s turn now to considering

More information

Lecture 4: An FPTAS for Knapsack, and K-Center

Lecture 4: An FPTAS for Knapsack, and K-Center Comp 260: Advanced Algorithms Tufts University, Spring 2016 Prof. Lenore Cowen Scribe: Eric Bailey Lecture 4: An FPTAS for Knapsack, and K-Center 1 Introduction Definition 1.0.1. The Knapsack problem (restated)

More information

A bad example for the iterative rounding method for mincost k-connected spanning subgraphs

A bad example for the iterative rounding method for mincost k-connected spanning subgraphs A bad example for the iterative rounding method for mincost k-connected spanning subgraphs Ashkan Aazami a, Joseph Cheriyan b,, Bundit Laekhanukit c a Dept. of Comb. & Opt., U. Waterloo, Waterloo ON Canada

More information

EXACT ALGORITHMS FOR THE ATSP

EXACT ALGORITHMS FOR THE ATSP EXACT ALGORITHMS FOR THE ATSP Branch-and-Bound Algorithms: Little-Murty-Sweeney-Karel (Operations Research, ); Bellmore-Malone (Operations Research, ); Garfinkel (Operations Research, ); Smith-Srinivasan-Thompson

More information

Vertex form of a quadratic equation

Vertex form of a quadratic equation Verte form of a quadratic equation Nikos Apostolakis Spring 017 Recall 1. Last time we looked at the graphs of quadratic equations in two variables. The upshot was that the graph of the equation: k = a(

More information

Approximation Algorithms and Hardness of Approximation. IPM, Jan Mohammad R. Salavatipour Department of Computing Science University of Alberta

Approximation Algorithms and Hardness of Approximation. IPM, Jan Mohammad R. Salavatipour Department of Computing Science University of Alberta Approximation Algorithms and Hardness of Approximation IPM, Jan 2006 Mohammad R. Salavatipour Department of Computing Science University of Alberta 1 Introduction For NP-hard optimization problems, we

More information

Improved Approximation for Single-Sink Buy-at-Bulk

Improved Approximation for Single-Sink Buy-at-Bulk Improved Approximation for Single-Sin Buy-at-Bul Fabrizio Grandoni and Giuseppe F. Italiano 2 Dipartimento di Informatica, Università di Roma La Sapienza, Via Salaria 3, 0098 Roma, Italy. grandoni@di.uniroma.it

More information

Continuity, End Behavior, and Limits. Unit 1 Lesson 3

Continuity, End Behavior, and Limits. Unit 1 Lesson 3 Unit Lesson 3 Students will be able to: Interpret ke features of graphs and tables in terms of the quantities, and sketch graphs showing ke features given a verbal description of the relationship. Ke Vocabular:

More information

Lessons from MIP Search. John Hooker Carnegie Mellon University November 2009

Lessons from MIP Search. John Hooker Carnegie Mellon University November 2009 Lessons from MIP Search John Hooker Carnegie Mellon University November 2009 Outline MIP search The main ideas Duality and nogoods From MIP to AI (and back) Binary decision diagrams From MIP to constraint

More information

Multicriteria approximation through decomposition

Multicriteria approximation through decomposition Multicriteria approximation through decomposition Carl Burch Sven Krumke y Madhav Marathe z Cynthia Phillips x Eric Sundberg { Abstract We propose a general technique called solution decomposition to devise

More information

variance of independent variables: sum of variances So chebyshev predicts won t stray beyond stdev.

variance of independent variables: sum of variances So chebyshev predicts won t stray beyond stdev. Announcements No class monday. Metric embedding seminar. Review expectation notion of high probability. Markov. Today: Book 4.1, 3.3, 4.2 Chebyshev. Remind variance, standard deviation. σ 2 = E[(X µ X

More information

NP-Completeness Review

NP-Completeness Review CS124 NP-Completeness Review Where We Are Headed Up to this point, we have generally assumed that if we were given a problem, we could find a way to solve it. Unfortunately, as most of you know, there

More information

The Maze Generation Problem is NP-complete

The Maze Generation Problem is NP-complete The Mae Generation Problem is NP-complete Mario Alviano Department of Mathematics, Universit of Calabria, 87030 Rende (CS), Ital alviano@mat.unical.it Abstract. The Mae Generation problem has been presented

More information