Distributed applications use predictions of network traffic to sustain their performance by adapting their behavior. The timescale of interest is applicationdependent and thus it is natural to ask how predictability depends on the resolution, or degree of smoothing, of the network traffic signal. To help answer this question we empirically study the one-step-ahead predictability, measured by the ratio of mean squared error to signal variance, of network traffic at different resolutions. A one-step-ahead prediction at a coarse resolution is a prediction of the average behavior over a long interval. We apply a wide range of linear and nonlinear time series models to a large number of packet traces, generating different resolution views of the traces through two methods: the simple binning approach used by several extant network measurement tools, and by waveletbased approximations. The wavelet-based approach is a natural way to provide multiscale prediction to applications. We find that predictability seems to be highly situational in practice—it varies widely from trace to trace. Unexpectedly, predictability does not always increase as the signal is smoothed. Half of the time there is a sweet spot at which the ratio is minimized and predictability is clearly the best. Also surprisingly, predictors that can capture non-stationarity and nonlinearity provide benefits only at very coarse resolutions. 1
|
1790
|
Congestion avoidance and control
– Jacobson
- 1988
|
|
1368
|
On the self-similar nature of Ethernet traffic
– Leland, Taqqu, et al.
- 1993
|
|
1101
|
lectures on wavelets
– Daubechies, Ten
- 1992
|
|
982
|
The Grid: Blueprint for a New Computing Infrastructure
– Foster, Kesselman
- 1999
|
|
507
|
Selfsimilarity through high-variability: statistical analysis of Ethernet LAN traffic at the source level
– Willinger, Taqqu, et al.
- 1997
|
|
425
|
The Network Weather Service: A Distributed Resource Performance Forecasting Service for Metacomputing
– Wolski, Spring, et al.
- 1998
|
|
318
|
TCP Vegas: End to End Congestion Avoidance on a Global Internet
– Brakmo, Peterson
- 1995
|
|
274
|
Comparison of Measurement-Based Admission Control Algorithms for Controller-Load Service
– Jamin, Shenker, et al.
- 1997
|
|
190
|
Dynamics of IP traffic: A study of the role of variability and the impact of control
– Feldmann, Gilbert, et al.
- 1999
|
|
171
|
Data networks as cascades: Investigating the multifractal nature
– Feldmann, Gilbert, et al.
- 1998
|
|
132
|
A multifractal wavelet model with application to network traffic
– Riedi, Crouse, et al.
- 1999
|
|
107
|
An Introduction to Long-Memory Time Series Models and Fractionally Di¤erencing
– Granger, Joyeux
- 1980
|
|
100
|
Fractional differencing
– Hosking
- 1981
|
|
89
|
Scheduling from the perspective of the application
– Berman, Wolski
- 1996
|
|
76
|
Statistical methods for data with long-range dependence
– Beran
- 1992
|
|
64
|
Threshold Models in Nonlinear Time Series Analysis
– Tong
- 1983
|
|
58
|
Predictive Dynamic Bandwidth Allocation for Efficient Transport of Real-Time VBR Video over ATM
– Chong, Li, et al.
- 1995
|
|
53
|
A Network Measurement Architecture for Adaptive Applications, Infocom
– Stemm, Katz, et al.
- 2000
|
|
48
|
Collecting network status information for networkaware applications
– Miller, Steenkiste
- 2000
|
|
46
|
Mobile network estimation
– Kim, Noble
|
|
43
|
An extensible toolkit for resource prediction in distributed systems
– Dinda, O’Hallaron
- 1999
|
|
40
|
Predictability analysis of network traffic
– Sang, Li
- 2000
|
|
37
|
Time series models for internet traffic
– BASU, MUKHERJEE, et al.
- 1995
|
|
35
|
Predicting the performance of wide area data transfers
– VAZHKUDAI, SCHOPF, et al.
- 2002
|
|
33
|
The architecture of the remos system
– Dinda, Gross, et al.
- 2001
|
|
33
|
Multiresolution approximation and wavelets
– MALLAT
- 1989
|
|
30
|
Online prediction of the running time of tasks
– DINDA
- 2002
|
|
23
|
A time series model of long-term NSFNET backbone traffic
– GROSCHWITZ, POLYZOS
- 1994
|
|
22
|
A new algorithm for measurement-based admission control in integrated services packet networks
– Casetti, Kurose, et al.
- 1996
|
|
21
|
Long-range dependence: Revisiting aggregation with wavelets
– ABRY, VEITCH, et al.
- 1998
|
|
21
|
Active Virtual Network Management Prediction Enhancement via
– Bush, F, et al.
|
|
11
|
On-line estimation of the parameters of long-range dependence
– ROUGHAN, VEITCH, et al.
- 1998
|
|
10
|
Time series models for Internet data traffic
– You, Chandra
- 1999
|
|
6
|
A time series model of long-term NSFNET backbone traffic
– GROSCHWITZ, POLYZOS
- 1994
|
|
5
|
Internet Traffic Archive. http://ita.ee.lbl.gov
– SIGCOMM
- 1999
|
|
5
|
Multi-resolution resource behavior queries using wavelets
– SKICEWICZ, DINDA, et al.
- 2001
|
|
4
|
A nonintrusive, wavelet based approach to detecting network performance problems
– HUANG, FELDMANN, et al.
- 2001
|
|
2
|
Long-Range Dependence: Theory and Applications. Birkhauser, 2002, ch. Self-similarity and long range dependence through the wavelet lens
– ABRY, FLANDRIN, et al.
|
|
2
|
Network traffic analysis, classification, and prediction
– QIAO, DINDA
- 2003
|
|
1
|
Virtualized audio: A highly adaptive interactive high performance computing application
– LU, DINDA
- 2002
|
|
1
|
Tsunami: A wavelet toolkit for distributed systems
– SKICEWICZ, DINDA
- 2003
|
|
1
|
Tsunami: A wavelet based toolkit for the analysis and dissemination of resource information
– SKICEWICZ, DINDA
- 2003
|
