Traffic Prediction on the Internet

1
Traffic Prediction on the Internet

• Anne Denton

2
Outline

• Paper by Y. Baryshnikov, E. Coffman, D.
  Rubenstein and B. Yimwadsana
• Solutions
• Time-Series prediction
• Our work for the KDD-cup 03

3
Time Series Prediction on the Internet

• By Y. Baryshnikov, E. Coffman, D. Rubenstein and
  B. Yimwadsana
• Adjustment to hot spots
• Avoiding degradation, even denial of service
• Can hot spots be predicted?
• Can predicted hot spots be avoided?

4
What are hot spots?

• Exceptionally large numbers of requests
• Spontaneous, short lifetime
• instant ramp up in traffic
• Only valid on long time scales
• Claim time scale for increase larger than time
  scale to react
• Why does increase take time?
• Passing on the word
• How good does a predictor have to be?
• Cost of missing a hot spot higher than
  aggregate cost of false alarms (similar to
  hurricane)

5
Examples

• Olympics (Nagano 98)
• Soccer World Cup (98)
• NASA (95)

6
What to do about hot spots?

• ltDetourgt The Columbia Hotspot Rescue Service A
  Research Plan
• E. Coffman, P. Jelenkovic, J.Nieh, and D.
  Rubenstein
• Approaches
• Deal ad hoc with high request
• Build a better network (expensive)
• Content delivery services
• Caching
• Extra bandwidth
• Suggested solution use available and
  underutilized resources

7
Hotspot Rescue Service

• Server-based approach
• Requires additional resources from server when
  necessary
• Resources provided by other members of Hotspot
  Rescue Service
• Peer-to-Peer approach
• Requires additional resources from client when
  necessary
• Caching

8
Four Phases

• Prediction (see rest of presentation)
• Server-based daemons
• P2P plug-ins
• Replication
• Server-based replication of objects
• P2P identified cached copies
• More advanced redistribution of traffic load
• Notification
• Modifications to DNS (Domain Name System)
• P2P system proactively announces hot objects and
  indicates alternative locations?
• Termination
• ltEnd of Detourgt

9
Tail of Distribution

• Requests per 10-second time slot
• X-axis number of hits per time slot
• Y-axis probability that that number of hits will
  be exceeded

10
Time Scales

• Prediction relies on correlation between values
  at different times
• Auto correlation function
• Predictability
• on time scales
• of 5-30 min

11
Prediction Algorithm

• Standard problem
• Signal processing
• Econometrics
• Internet traffic
• Particularly bursty
• Simplest model
• Linear extrapolation

12
Structure of Prediction Algorithms

• Traffic observation
• of requests in time unit (t-1,t
• Usually 1s
• Prediction window
• Duration Wp ? 0
• Advance notice ?
• Prediction at time t
• Mapping of observations in t-Wp,t to a number
  pt ? 0 of requests predicted in interval
• t?, t?1 that is ? units in the future

13
Linear Prediction

• Linear Fit Least squares linear fit
• pt ft(t?) with
• ft(s) at sbt
• Minimizing
• Performance O(WT)
• W Window size
• T uptime duration
• Problems
• Prediction window size must match burstiness
  parameters governing request flow

14
Results

• Depends on properties of auto-correlation
  function

15
Conclusions of Paper

• Build a load-based taxonomy of web server traffic
• Depends on technological, sociological, and
  psychological factors
• Look for quantification of basic patterns
  reflecting behavior
• Do we agree ???
• Why cluster when we can classify!!

16
Our Approach

• Normally time series prediction uses only data in
  that time series
• We use similarity to other instances
• E.g., other web sites
• Model-free
• Weighted Nearest Neighbor approach
• Problem
• How integrate time?

17
Typical Nearest Neighbor Classification /
Regression

• R(A1, , An, C)
• Attributes Ai
• C class label (classification)
• or continuous variable (regression)
• Based on distance function on Ai
• K nearest neighbors
• Neighbors within a range
• Use kernel function to weight closer ones higher

18
Weighting of Attributes

• Some attributes are more important than others
• Apply scaling to space
• Optimize weights through
• Hill-climbing
• Genetic Algorithm
• How does this generalize to a time-series?

19
Our Answer

• Identify relevant sections in the time series
• E.g. times with already high download rates
• Well call each relevant section a prediction

20
Predictions

• Each prediction contains information about
• The nature of the time series
• The time instance in question, i.e. the history
  of requests
• The actual change in requests
• Make a table of predictions
• Leads to a relation just as standard
  classification / regression setting

21
Data Set

• Paper citations in e-print ArXive
• Background KDD-cup 03
• Predict the change in citations in successive
  3-month periods
• Only consider periods with at least 6 citations
• Evaluation L1 distance (Manhattan distance)
  between predicted and real difference
• Very close match between citation history and
  request history
• Predict change in requests
• Only consider periods that already show large
  number of requests

22
Attributes of a Prediction

• Quantitative attributes
• Number of citations in window
• Gradient of citations in window
• Aggregate number of citations up to and through
  window (assume finite time series)
• Attribute values given by time series
• Keyword occurrences
• Author
• Number of revisions of papers
• Maximum time interval between revisions
• Country of origin
• Format

23
Similarity Function

• Common kernel-function
• What worked better

24
Plot of Similarity Function
25
Accuracy

• No linear extrapolation data available
• Could lead to negative citations
• Comparison
• Default prediction No change 1851
• Very simple model (decrease by 0.3 in 3 months)
  1532
• Prediction based on average of time series
  (synchronized at first non-0) 1593
• Prediction based on quantitative attributes 1465
• Full prediction (prelimiary) 1357
• Weight optimized (very preliminary) reduction
  1414 -gt 1391

26
Results
27
Conclusions

• Method works well for citation prediction
• Yet to be tested for hot-spot prediction