1. 1 Distributed Partial Information Management (DPIM) for Survivable Networks Dahai Xu

2. 2 Content Basic Concepts of Protection & Restoration Previous Work on Shared Path Protection Proposed DPIM Schemes what partial info to maintain and how? how a connection is routed under distributed control and with partial info? how a connection is routed under distributed control and with partial info? how distributed signaling is done and bandwidth (BW) allocated/deallocated? how distributed signaling is done and bandwidth (BW) allocated/deallocated? A heuristic based on Potential Backup Cost A heuristic based on Potential Backup Cost

3. 3 Protection Path Protection Link Protection Advantages & Disadvantages

4. 4 Path Protection Use more than one path to guarantee the data be sent successfully Dedicated Path Protection Shared Path Protection

5. 5 Dedicated Path Protection 1+1 Protection Point-to-Point Protection & Mesh Network Protection

6. 6 1+1 Protection

7. 7 Mesh Network Protection

8. 8 Shared Path Protection 1:1 Protection 1:N Protection

9. 9 Link Protection Use an alternate path if the link failed Dedicated Link Protection: not practical Shared Link Protection: practical It may fail when a node fails

10. 10 Advantages & Disadvantages of Protection Simple Quick: Do not require much extra process time Usually can only recover from single link fault Inefficient usage of resource

11. 11 Restoration Path Restoration Route can be computed after failure Link Restoration Path is discovered at the end nodes of the failed link More practical than path restoration Advantages & Disadvantages of Restoration Usually can recover from multiplex element faults More efficient usage of resource Complex Slow: require extra process time to setup path and reserve resource

12. 12 Characteristic: Protection -- the resource are reserved before the failure, they may be not used; Restoration -- the resource are reserved and used after the failure Route: Protection -- predetermined; Restoration -- can be dynamically computed Resource Efficiency: Protection -- Low; Restoration -- High Comparison between Protection & Restoration

13. 13 Time used: Protection -- Short; Restoration -- Long Reliability: Protection -- mainly for single fault; Restoration -- can survive under multiplex faults Implementation: Protection -- Simple; Restoration -- Complex Comparison between Protection & Restoration (Cont ’ )

14. 14 Offline Routing Arrange a set of traffic flows Integer Linear Programming(ILP) to get optimal results Heuristic Algorithms Relaxation of ILP Simulated Annealing - A stochastic hill-climbing heuristic search method. (Explore a larger area in the search space without being trapped in local optimal) Genetic Algorithm: Evolves the current population of “good solutions” toward the optimality by using carefully designed crossover and mutation operators. Tabu search

15. 15 Online Routing of Bandwidth Guaranteed Online routing, bandwidth guaranteed path with simultaneous protection path Metrics Unlimited Link Capacity Bandwidth Consumption Limited Link Capacity Connection drop/block probability Profit / Revenue

16. 16 Assumption Two connections whose active paths are completely link disjoint can share backup Bandwidth (BBW). The objective of the algorithm is to exploit this BBW sharing to e.g., reduce the total amount of bandwidth (TBW) consumed by the connections.

17. 17 Information for Routing The amount of BBW sharing depends on the information available to the routing algorithm. Three important cases to be considered. No Information on how existing connections are routed Complete Per-flow/Aggregate Information Partial Aggregate Information

18. 18 No Sharing (NS) Only know the residual (available) bandwidth on each link Residual bandwidth = Link capacity - Reserved active bandwidth (ABW) - Reserved backup bandwidth (BBW) Can be obtained from OSPF Extensions or IS­ IS Extensions Only the total used bandwidth is known (active + backup) Can not share BBW, thus waste resources.

19. 19 Sharing with Complete Information (SCI) Know routes for the active and backup paths of all current connections. May have too much information to maintain. O(LQ). L is the average path length, Q is the number of existing connections. Permits the best sharing and provides a Performance upper-bound

20. 20 Partial Information for Routing Know some aggregated information of each link Two schemes SPI (Sharing with Partial Information): Centralized control, knows BBW and ABW on each/every link DPIM (Distributed Partial Information Management): Distributed control, each ingress edge (source) node decides the routes.

21. 21 Notations (I)

22. 22 Notations (II)

23. 23 No Sharing (NS) Remove links R e < w Determine two link disjoint paths for active/backup Formulation: standard network flow problem each link has unit cost and unit capacity s supply two units, d demand two units minimum cost flow algorithm can be used

24. 24 Linear Programming for SCI (I) For new request (s, d, w), the least cost of using a on AP and b on BP The cost of using e on BP (1)

25. 25 Linear Programming for SCI (II) Objective Constraints

26. 26 SPI In SCI, can be calculated from per-flow information. Need maintain per-flow information. Not scalable. In SPI,is not known, only is known Same objective and constraints as in SCI Further improvement to be discussed in DPIM

27. 27 Survivable Routing (SR) Distributed control with complete but aggregated information. Every edge node essentially maintains a matrix of for all links a and b Uses the active path first (APF) heuristic instead of ILP formulation Remove links whose R e <w (temporarily) Find a shortest path as AP Put back temporarily removed links, remove AP links, calculate backup cost using Eq. (1) Find a shortest (cheapest) path as BP

28. 28 Successive SR (SSR) After is updated as a result of setting up a new connection, some existing BPs may change (route and the amount of additional BBW reserved) Such changes may in turn trigger changes to other existing BPs until an equilibrium state is reached Achieve a better BBW sharing, but with a high signaling and control overhead

29. 29 RAFT RAFT: Resource Aggregation for Fault Tolerance Each node maintains fault management table (FMT), which list AP or BP flow on each link e. FMT must be updated each time a request initiates or terminates AP and BP route are node-disjoint by using shortest path algorithm firstly A request is accepted only if the bandwidth requirement is available on all the links on its AP and BP, otherwise it is rejected.

30. 30 Doshi’s Each node maintains a link capacity control table (LCCT) for each local link Source nodes using Content-lock mechanism to avoid multiple demands deadlock. BP route search: Distributed breadth-first search (BFS) over a residual network In BFS, it first query the residual spare capacity in LCCT, only use the link if the link has sufficient capacity If a route is found, the source node stores it as the restoration route for the demand. If fail to find the BP route, the capacity optimization procedure is activated by changing previous BP routes

31. 31 Su’s Each node maintains “bucket”-based link state (equivalent to ) The amount of link states is proportional to the number of failure/link, not the number of light paths AP and BP are optimized separately. AP are assumed to using minimum-hop paths, BP are optimized to reduce the wavelength redundancy The “width” of link l with respect to a failure event k* is defined as the normalized difference between the maximum bucket height and the bucket corresponding to link failure k*, which indicates the sharing capacity of links.

32. 32 By using Bellman-Ford algorithm to identify the widest path between the end nodes of the protected link, the path that offer the most sharing. In the event that there are more than one such path candidates, the one that traverses the lease number links with width 0 was selected Su’s (Cont’)

33. 33 DPIM-SAM Distributed Partial Information Management Edge node maintains (and exchanges) non-local information: for each link e. (O(E) information) Each node also maintains profiles of ABW and BBW for each local link e. (O(E) information)

34. 34 Path Determination This estimated BBW may not be minimal Using ILP, or APF to find AP and BP DPIM-M-A: APF with Minimal BBW Allocation

35. 35 Distributed Signaling Minimal BBW Allocation Maintaining Partial Information on AP and BP Send AP Set-up packet containing BP to the nodes along AP, each node having an outgoing link e in AP updates Similar way to update

36. 36 Minimal BBW allocation

37. 37 Connection Release Can’t be done efficiently in SPI AP Tear-Down and BBW Deallocation. Update P B e and release bw.

38. 38 Network Topology

39. 39 Performance Evaluation Traffic Types Incremental traffic (Established connection lasts forever) Dynamic traffic (with connection durations) Performance Metrics Unlimited Link Capacity Bandwidth Saving (Ratio): upper bound 50% Limited Link Capacity Connection drop/block probability Total Earning (Ratio) : Earning Rate matrix (independent of traffic load)

40. 40 Simulation Results Average Bandwidth Saving Ratio Total Earning Ratio

41. 41 Active Path First with Potential Backup Cost (APF-PBC) Challenges Integer Linear Programming (ILP) based approaches are notoriously time consuming Guarantee minimal allocation of TBW for each request, but do not guarantee an optimal result for all requests. Active path first (APF) can only achieve sub-optimal results: Does not consider the potential cost along the BP when selecting the AP

42. 42 Main idea of APF-PBC Also uses Active Path First In selecting Active Path, Each capable link a will be assigned a cost We use as the potential backup cost (and try to minimize TBW). Intuition: PBC increases with w and Can apply to SCI and DPIM-SAM (which determine backup cost and BP differently)

43. 43 Potential Backup Cost - Derivation is derived based on the statistical analysis of experimental data. (SCI-ILP) for the 15- node network, infinite link capacity) challenge: but do not know which link b to be used to backup link a, let alone B b and solution: guess the (weighted average) value of B b (call it x) and (call it s)

44. 44 Derivation based on statistical analysis of B b Distribution of B b /M Distribution of B b /M  (w,s,M) is the expected value of  a (w) when s is fixed. Guess the distribution of and calculated the weighted average value of  (w,s,M) over all s to obtain  a (w)distribution of

45. 45 Distribution of B b /M

46. 46 Graph of  (w,s,M) & approximation Integral (curves) from adaptive Lobatto quadrature Approximation (line-fitting Y=c 1 X+c 2 )

47. 47 Cumulative distribution function of

48. 48 Graph of

49. 49 Approximation of  a (w) Distribution of Effect of constants c and  on performance of APF-PBC Effect of constants c and  on performance of APF-PBC

50. 50 Distribution of

51. 51 Effect of constants c and  on performance of APF-PBC

52. 52 Bandwidth consumed after 500 demands

53. 53 Total earning after 500 demands

54. 54 Simulation Results -PBC Average Bandwidth Saving Ratio Total Earning Ratio

55. 55 Summary On-line Shared path protection (need to extend to other schemes) Amount of information (Complete/Partial) affects BBW sharing May use ILP or APF-based heuristics Proposed a DPIM scheme for a distributed, partial / aggregated information management (including signaling for path set-up/tear-down) Proposed a potential cost heuristic, which runs faster and better than ILP

56. 56 Summary II Have also extended to cases with unprotected (UP) and pre-emptable (PE) connections UP: use just one path similar to an AP (i.e., no BP); affected if (and only if) the path breaks. PE: unprotected and may be affected even if a failure does not break its path A PE may use the existing BPs/BBW to carry low- priority traffic in fault-free situations A PE is similar, but not identical to a BP: can share BBW with other BPs, but cannot share with other PE The idea of potential cost can also be applied to solving other joint optimization problems with heuristics

57. 57 Reference