1 / 37

On Scalable Measurement-driven Modeling of Traffic Demand in Large WLANs

On Scalable Measurement-driven Modeling of Traffic Demand in Large WLANs. Maria Papadopouli 1,2. 1 Foundation for Research & Technology-Hellas (FORTH) & University of Crete 2 University of North Carolina at Chapel Hill.

rgina
Download Presentation

On Scalable Measurement-driven Modeling of Traffic Demand in Large WLANs

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. On Scalable Measurement-driven Modeling of Traffic Demand in Large WLANs Maria Papadopouli1,2 1Foundation for Research & Technology-Hellas (FORTH) & University of Crete 2 University of North Carolina at Chapel Hill 1IBM Faculty Award, EU Marie Curie IRG, GSRT “Cooperation with non-EU countries” grants

  2. Wireless landscape • Growing demand for wireless access • Mechanisms for better than best-effort service provision • Performance analysis of these mechanisms • Majority of studies make high-level observations about traffic dynamics in tempo-spatial domain • Models of network &user activityin various spatio-temporalscales are required

  3. User B Wireless infrastructure disconnection Internet Router Wired Network AP3 Switch Wireless Network User A AP 1 AP 2

  4. 1 2 3 0 Wireless infrastructure Internet disconnection Router Wired Network Switch AP3 Wireless Network User A AP 1 AP 2 roaming roaming User B Associations Flows Packets

  5. Modelling objectives • Important dimensions on wireless network modelling • user demand (access & traffic) • topology (network, infrastructure, radio propagation) • Structures that are well-behaved, robust, scalable & reusable • Publicly available analysis tools, traces, & models

  6. Internet disconnection Wired Network Router Switch AP3 Wireless Network User A AP 1 AP 2 Events User B Session 1 2 3 0 Association Flow Arrivals t1 t2 t3 t4 t5 t6 t7 time

  7. Wireless infrastructure & acquisition • 26,000 students, 3,000 faculty, 9,000 staff in over 729-acre campus • 488 APs (April 2005), 741 APs (April 2006) • SNMP data collected every 5 minutes • Packet-header traces: • 175GB (April 2005), 365GB (April 2006) • captured on the link between UNC & rest of Internet via a high-precision monitoring card

  8. Our models • Session • arrival process • starting AP • Flow within session • arrival process • number of flows • size Captures interaction between clients & network Above packet level for traffic analysis & closed-loop traffic generation

  9. Our parameters and models

  10. Related modeling approaches • Hierarchical modeling by Papadopouli [wicon ‘06] Parameters: Session & in-session flow: • Time-varying Poisson process for session arrivals • biPareto for in-session flow numbers & flow sizes • Lognormal for in-session flow interarrivals • Flow-level modeling by Meng [mobicom ‘04] • No session concept, flow interarrivals follow Weibull • AP-level over hourly intervals •  Larger deviation from real traces than our models

  11. Number of Flows Per Session

  12. Related modeling approaches (cont’d) Objective Scales

  13. Main research issues • Hierarchical modeling traffic workload AP-level vs. network-wide Other spatio-temporal levels ? • Model validation @ different spatial scales using data from different periods • Scalability, reusability, accuracy tradeoffs

  14. Hourly session arrival rates

  15. More active web browsing behavior Session-level flow variation Broadvariation of the in-session number of flows per building-type distribution Number of flows in a session (k)

  16. Session-level flow size variation Mean flow size f (bytes)

  17. Session-level flow related variation In-session flow interarrival can be modeled with same distribution for all building types but with different parameters Mean in-session flow interarrival f

  18. Starting building & “roaming” Small % of building-roaming flows Little dependence on what kind of building a session is initiated Number of visited bldgs x

  19. Model validation Simulations: synthetic data vs. original trace • Metrics: variables not explicitly addressed by our models • aggregate flow arrival count process • aggregate flow interarrival (1st & 2nd order statistics) • Increasing order of spatial aggregation AP-level, building-level (bldg), building-type-level (bldg-type), network-wide • Different tracing periods (April 2005 & 2006)

  20. Simulations Produce synthetic data based on aforementioned models • Synthesize sessions & flows for simulations • Session arrivals are modeled after hourly bldg-specific data • Flow-related data: bldg (day, trace), bldg-type, network-wide

  21. Simplicity at the cost of higher loss of information Number of flows per session

  22. Number of aggregate flow arrivals

  23. Autocorrelation of flow interarrivals

  24. Aggregation in time-dimension may cancel out the benefit of getting higher spatial resolution Flow interarrivals time

  25. Conclusions Multi-level parametric modelling of wireless demand • Network-wide models: • Time-varying Poisson process for session arrivals • biPareto for in-session flow numbers & flow sizes • Lognormal for in-session flow interarrivals • Validation of models over two different periods • Same distributions apply for modeling at finer spatial scales building-level, groups of buildings with similar usage • Evaluation of scalability-accuracy tradeoff

  26. UNC/FORTH web archive  Online repository of models, tools, and traces • Packet header, SNMP, SYSLOG, signal quality http://netserver.ics.forth.gr/datatraces/  Free login/ password to access it Joint effort of Mobile Computing Groups @ FORTH & UNC  maria@csd.uoc.gr

  27. Appendix

  28. Related research Modeling traffic in wired networks • Flow-level • several protocols (mainly TCP) • Session-level • FTP, web traffic • session borders heuristically defined by intervals of inactivity Modeling traffic in wireless networks • Flow-level modeling by Meng [mobicom04] • No session concept, flow interarrivals follow Weibull • Modelling flows to specific APs over one-hour intervals  Does not scale well  Larger deviation from real traces than our models

  29. Flow interarrival time [Hinton-James

  30. Hourly number of flow arrivals [Hinton-James

  31. Autocorrelation of flow interarrivals [Hinton-James

  32. HT James

  33. McColl

  34. Our models 2/2 N: #sessions between and

  35. Related work in wireless traffic modeling • Over hourly intervals at AP-level • Captures finer spatial detail required for evaluating network functions with focus on AP-level (e.g., load-balancing, admission control) • Does not scale for large infrastructures • Data do not always amenable to statistical analysis • Infrastructure-wide • Models amenable to statistical analysis • Concise summary of traffic demand at system-level • Fails to capture finer spatial detail required for evaluating network functions with focus on AP-level

More Related