1. Sterbenz, et al. ITTC Cross-Layering and Metrics ResiliNets and ResumeNet 07 October 2009 James P.G. Sterbenz* † Джеймс Ф.Г. Стербэнз 제임스 스털벤츠 司徒傑莫 David Hutchison, Abdul Jabbar, Radovan Brunčák Justin P. Rohrer, Egemen Çetinkaya *Department of Electrical Engineering & Computer Science Information Technology & Telecommunications Research Center The University of Kansas † Computing Department, Infolab 21 Lancaster University jpgs@ittc.ku.edu http://www.ittc.ku.edu/~jpgs http://wiki.ittc.ku.edu/resilinets © 2009 Sterbenz

2. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet2 Cross-Layering and Metrics Abstract Cross-layer theory: mechanisms and algorithms to control state transactions among several state machines within required level Metrics: a notation to describe state transactions of particular state machine

3. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet3 ResiliNets Motivation for Cross-Layering D 2 R 2 strategy motivates… Principles for resilient networking, e.g. –Adaptability –State management –Multilevel –Context awareness –Autonomic (self-optimization) Requires cross-layering Theory, algorithms, mechanisms and methodologies to support resilience

4. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet4 Knobs and Dials Formalism Set of all knobs  = K  k –union of out-of-band K and in-band k Set of all dials  = D  d –union of out-of-band D and in-band d Set of knobs or dials between layers L i and L j –where i and j are either numbers, e.g. {1,1.5,2,2.5,3,4,7} or designators, e.g. {HBH, net, PoMo, E2E, app, …} – K i  j =   K  K K i  j –vertical when i≠j ; horizontal when i=j Individual knob or dial between layers L i and L j –K i  j (d) where d is a descriptor, e.g. BER

5. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet5 Knobs and Dials Formalism A protocol instance has –state at time t of s(t) –context at time t of c n (t) At L n, state is –s(t+1) = f (  n +1  n,  n  n –1, s(t), c n ) SM k n +1  n LnLn K n +1  n d n +1  n D n +1  n k n  n –1 d n  n –1 K n  n –1 D n  n –1 K,D n  n k,d n  n cncn s(t)s(t)

6. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet6 Cross-Layer Model Protocol Instance Model Node state –state machine –memory Node inputs/outputs –vertical data –p2p virtual data –context –horizontal signalling –cross-layer signalling SM E2EE/HpayloadHBH

7. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet7 ResiliNets Cross-Layer Model E2E and HBH layers HBH E2E E/HpayloadHBH E2E context HBH context E2E context HBH context physical environment dials/knobs application knobs/dials cross-layer in-band signalling horizontal explicit signalling vertical explicit signalling

8. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet8 End-to-End Communication Knobs and Dials Knobs and dials between upper layers and PoMo –support heterogeneous subnetworks e.g. lossy wireless vs. reliable wired –explicit signalling of path diversity and multipath geographic location of realms, nodes, channels Knobs  Dials  Layer application E2E transport PoMo internetwork network realm HBH link service characteristics path char., geography realm characteristics link characteristics service class reliability mode PoMo knobs, FD, motiv. realm oper. parameters link type and coding error control type/strength

9. Sterbenz, et al. ITTC Goal Make progress in understanding fundamental principles of cross- layering Deliver cross-layer theory –Cross-layer formalisms State space, variables, operations, transitions –Cross-layering calculus Cost and benefit calculation Theory to compare cross-layer mechanisms, reason about costs and benefits Mechanisms build up on standard building blocks (in-baud vs. out-of- baud, open vs. control loop, E2E vs. HBH) –Cross-layer evaluation model Dynamics of the cross-layer system, avoid instabilities, oscillations, etc. 07 October 20099

10. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet10 Case study: error control Functional alternatives Nnone Oopen loop (FEC) Cclosed loop (ARQ) S&W, GB-N, SelRep Location –HBH –E2E App requirements –unreliable –quasi-reliable –reliable N one O FEC C ARQ N one O FEC C ARQ E2E HBH

11. Sterbenz, et al. ITTC Approaching problem Cross-layer control: between E2E and HBH error control One hop –Select error control (ARQ,FEC,NONE) –Setup strength –Local decision only –Inputs: state, context, knob, dials Hop-by-Hop –Calculating dials End-to-End –Selecting error control and setup strength –Calculating knobs Cost contrasting methodology 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet11

12. Sterbenz, et al. ITTC Selecting ER control: Reasoning about costs Variables: Capacity, Speed, BER, BER interval, dynamics of BER Capacity: High => FEC Low => ARQ Latency:High => ARQ Low => FEC BER: HIGH => FEC, ARQ LOW => FEC, ARQ BER interval:HIGH => FEC, ARQ LOW => FEC, ARQ BER dynamics:HIGH =>ARQ LOW => FEC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet12

13. Sterbenz, et al. ITTC Cost-Contrasting Methodology 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet13 BER interval BER dynamics HIGH LOW ARQFEC BER

14. Sterbenz, et al. ITTC Selecting ER control: Reasoning about benefits Guarantee: –HIGH because there is feedback => ARQ –LOW because there is not feedback => FEC However, it can be HIGH, if estimation of the link characteristics is good. However we need to deal with Probabilities Latency –Global remedy (E2E) is slow –Local remedy (HBH) is quick 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet14

15. Sterbenz, et al. ITTC Setup strength p – error rate, proportional to BER s – secure constant – depends on state, context (can go even to negative numbers) m – mode, (unreliable ->0, quasi-reliable -> 0.5, reliable -> 1)? 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet15

16. Sterbenz, et al. ITTC Dial calculation: possible approaches Max BER value Significant BER change Average BER value Number of unmanaged packets 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet16

17. Sterbenz, et al. ITTC Knob calculation: possible approaches Knob –Reliable –Quasi-reliable –Unreliable Calculation? 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet17

18. Sterbenz, et al. ITTC Dynamics of cross-layer system Knob can be change changed in real-time How to describe the dynamics of the underline system? How can we describe the state transitions? How can control theory be applied? 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet18

19. Sterbenz, et al. ITTC Control theory in cross-layering? 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet19

20. Sterbenz, et al. ITTC Control theory in cross-layering? Dynamics of the cross-layer system? How knobs can change state of the underline system ? 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet20

21. Sterbenz, et al. ITTC Cross-layering calculus Decomposition? Towards a cross-layer calculus: 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet21

22. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet22 Resilience and Heterogeneity Evaluation Methodology: Simulation Resilience strategy and principles Postmodern Internet Heterogeneity Example realms –WDTN –highly mobile airborne ad-hoc networking Evaluation methodology –simulation –experimentation

23. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet23 Evaluation Methodology Flexible and Realistic Topology Generation Hierarchical topology generation –evaluation of PoMo mechanisms –network engineering for resilience Level 1: backbone realms –nodes distributed based on location constraints –links generated using various models under cost constraints Level 2: access network realms –distributed around backbone nodes –access network connectivity: ring, star, mesh Level 3: subscribers –distributed around access network node

24. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet24 Evaluation Methodology Challenge Simulation Module Separate challenge from network simulation Simulate challenges to any network over time interval –natural disaster destroy network infrastructure (e.g. Katrina) or large-scale grid failure (e.g. Northeast US 2003) –attack: {node|link} down, wireless link attenuated  

25. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet25 Evaulation Methodology Example: Sprint Actual Topology

26. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet26 Evaluation Methodology Example: Sprint Synthetic Fragile Topology

27. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet27 Evaluation Methodology Challenge Simulation Module # of nodes down aggregate packet delivery ratio

28. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet28 Evaluation Methodology Challenge Simulation Module KU-CSM Challenge Simulation Module –challenge specification describes challenge scenario –network coordinates provide node geo-locations –adjacency matrix specifies link connectivity –input to conventional ns-3 simulation run –generates trace to plot results adjacency matrix.txt 0 0 1 0 simulation description.cc challenge specification.txt node coordinates.txt ns-3 simulator (C++) plotted trace KU-LoCGen

29. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet29 Resilience Measure Towards a Resilience Metric Need: analyse and understand level of resilience Problem: how to measure –ideal:  = [ 0,1 ] –but too many metrics to combine into a single number perhaps a small set of objective functions –need to separate service spec. from operational parms. Model resilience as two-dimensional state space –operational state –service level

30. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet30 Resilience Metrics Multilevel Approach Resilience defined at any layer boundary –operational state describes system below layer boundary –service states represents behaviour above boundary Operational range arbitrarily divided into 3 regions –normal operation –partially degraded –severely degraded Service delivered arbitrarily divided into 3 regions –acceptable service –impaired service –unacceptable service

31. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet31 Resilience Metrics States State of the system represented by ( N, P ) tuple –multivariate operational state N –multivariate service state P Initial state described by –acceptable service –during normal operations Challenges perturb N and P –resilience trajectory

32. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet32 Resilience State Space Operational Resilience Normal Operation Partially Degraded Severely Degraded Operational State N S Operational resilience –minimal degradation –in the face of challenges Resilience state –remains in normal operation

33. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet33 Resilience State Space Service Resilience Normal Operation Partially Degraded Severely Degraded Acceptable Impaired Unacceptable Operational State N Service Parameters P S Service resilience –acceptable service –in the face of degraded operation Resilience state –remains in acceptable service S

34. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet34 Resilience State Space Resilience Trajectories Choose scenario –network –application Metrics –choose –aggregate Observe –under challenge Unacceptable Impaired Acceptable P [delay, pkt delivery ratio] Normal operation Partially degraded Severely degraded S5S5 S1S1 S4S4 p 1 > 10 sec p 2 > 0.85 p 1 < 10 sec p 2 > 0.50 p 1 < 1 sec p 2 < 0.50 n 1 ≤ 1.5ρ 0 n 2 ≥ 3.0 n 1 ≤ 3ρ 0 n 2 ≥ 1.0 n 1 > 3ρ 0 n 2 < 1.0 N [load, node degree] S3S3 S2S2

35. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet35 Resilience Evaluation Basic Approach Define service boundary and metrics –boundary B ij between two adjacent layers L ij – k operational metrics characterize network below B ij –N ij = {N 1, N 2, … N k } – l service parameters define service across B ij –P ij = {P 1, P 2, … P k } Evaluate resilience –operational and service space divided in to regions –resilience R ij at the boundary B ij is evaluated as …. … the transition of the network through the state space – R ij = f (N ij,P ij )

36. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet36 Resilience Evaluation Multilevel Mapping of service to operations –service parameters become operations at layer above N i+1, j+1 =P ij –resilience can be evaluated at any arbitrary layer boundary –resilience as seen by the application is at the L 67 boundary

37. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet37 Resilience Evaluation Time Scale Steady state analysis –long term view of resilience –theoretical evaluation or simulation study to understand impact of perturbations on operations to service being provided –best, worst, and mean case type of studies are possible Transient analysis –based on instantaneous state of the network –plot service parameters vs. operational metrics in real time –resilience is characterized by state transitions –shows the impact of resilience mechanisms defense, remediation, and recovery

38. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet38 Resilience Evaluation Preliminary Results Resilience study at boundary B 23 –operational state of network is characterized by vertices, edges, and edge failures –service provided across the boundary is a connected graph –steady state analysis based on simulations Sprint case study –operational metric: link failures f (fixed nodes and links) –service parameters: (preliminary, proof of concept) avg. node degree d relative size of largest connected component | c max | –normalised to number of nodes in graph

39. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet39 Resilience Evaluation Preliminary Results: Sprint Case Study 1000 runs Conventional plot –limited to 1 metric/axis Resilience state space –discrete approach –map into N, P regions

40. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet40 Resilience Evaluation Preliminary Results: Sprint Case Study normal l = 0 partially degraded 0 < l ≤ 5 severely degraded 5 < l ≤ 68 unacceptable d < 2 & | c max | < 0.95 impaired 4 > d ≥ 2 & 1 > | c max | ≥ 0.95 acceptable d ≥ 4 & | c max | = 1 # of link cuts average node degree, component size

41. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet41 Resilience Evaluation Multilevel Operational vs. Service State Operational State N i Service Parms. P i+1 Operational State N i+1  P i+1 B i,i+ 1 B i+1,i+ 2 Service Parms. P i+2

42. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet42 Resilience Evaluation Multilevel Operational vs. Service State LayerServiceOperational State Application (7) application behaviourE2E performance (delay, goodput, …) E2E Transport (4) E2E data transferstable E2E path Inter-realm path (3.5 PoMo) E2E inter-realm path (routing and forwarding) stable intra-realm path Subnetwork path (3) e2e intra-realm path (routing and forwarding) topology Topologyconnected graph#link cuts, nodes out HBH (2) stable link HBH data transfer channel conditions

43. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet43 Resilience Evaluation Preliminary Results: Sprint Case Study Conclusions –shows an example of resilience characterization at L 23 –best case scenario, the topology is highly resilient –worst case, single link failure results in unacceptable service Future work –evaluate resilience R 34 at boundary B 34, d and | c max | become the operational metrics at B 34 –compare multiple topologies against each other Sprint, GÉANT2, synthetic topologies generated with KULocGen –conduct transient analysis see the impact of D 2 R 2

44. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet44 End

45. Sterbenz, et al. ITTC 07 October 2009Resilience, Survivability, Heterogeneity in Postmodern Internet45 End-to-End Communication Knobs and Dials in PoMo Header In band: knobs and dials in data packet Explicit signalling: knobs and dials in signalling packet realm E2E realm instrumentation dials/knobs application knobs/dials cross-layer in-band signalling horizontal explicit signalling vertical explicit signalling