© The Institution of Engineering and Technology
Packet level measurement is now routinely used to evaluate the loss and delay performance of broadband networks. In active measurement, probe packets provide samples of the loss and delay and from these samples the performance of the traffic as a whole can be deduced. However this is prone to errors: inaccuracy due to taking insufficient samples, self-interference due to injecting too many probe packets, and possible sample-correlation induced bias. In this paper we consider the optimisation of probing rate by treating all measurements as numerical experiments which can be optimally designed by using the statistical principles of design of experiments. We develop an analytical technique that quantifies an overall utility function associated with: (i) the disruption caused per probe packet, (ii) the bias and (iii) the variance as a function of the probing (sampling) rate. Our numerical results show that the optimal probing rate depends strongly on what parameter the network engineer seeks to measure.
References
-
-
1)
-
20. Baccelli, F., Machiraju, S., Veitch, D., Bolot, J.: ‘On optimal probing for de-lay and loss measurement’. Proc. Seventh ACM SIGCOMM Conf. on Internet Measurement, ACM, 2007, pp. 291–302.
-
2)
-
23. Sun, S., Xiao, W.: ‘Optimal linear estimators for systems with multiple random measurement delays and packet dropouts’, Int. J. Syst. Sci., 2013, 44, (2), pp. 1–13, . (doi: 10.1080/00207721.2013.843732).
-
3)
-
12. Basawa, I., Prabhu, N.: ‘Large sample inference from single server queues’, Queueing Syst. Theory Appl., 1988, 3, (4), pp. 289–304. (doi: 10.1007/BF01157852).
-
4)
-
R. Wolff
.
Poisson arrivals see time averages.
Oper. Res.
,
2 ,
223 -
231
-
5)
-
22. Baccelli, F., Machiraju, S., Veitch, D., Bolot, J.: ‘Probing for loss: the case against probe trains’, Commun. Lett., IEEE, 2011, 13, (99), pp. 590–592. (doi: 10.1109/LCOMM.2011.030911.101495).
-
6)
-
5. Parker, B.M., Gilmour, S.G., Schormans, J.: ‘Measurement of packet loss probability by optimal design of packet probing experiments’, IET Commun. Special Edition on Simulation, Analysis and Measurement of Broadband Network Traffic, 2009, 3, (6), pp. 979–991. .
-
7)
-
14. Basawa, I., Bhat, U., Lund, R.: ‘Maximum likelihood estimation for single server queues from waiting time data’, Queueing Syst., 1997, 24, pp. 155–167 (doi: 10.1007/BF01149084).
-
8)
-
4. Atkinson, A., Donev, A., Tobias, R.: ‘Optimum experimental designs, with SAS’ (Oxford University Press, 2007).
-
9)
-
28. Aigner, D.: ‘Parameter estimation from cross-sectional observations on an elementary queuing system’, .
-
10)
-
7. Jenkins, J.: ‘The relative efficiency of direct and maximum likelihood estimates of mean waiting time in the simple queue M/M/l’, J. Appl. Probab., 1972, 9, (2), pp. 396–403 (doi: 10.2307/3212807).
-
11)
-
11. Billingsley, P.: ‘Statistical inference for markov processes’ (The University of Chicago Press, 1962).
-
12)
-
F. Baccelli ,
S. Machiraju ,
D. Veitch ,
J. Bolot
.
The role of PASTA in network measurement.
Comput. Commun. Rev.
,
4 ,
231 -
242
-
13)
-
9. Reynolds, J.: ‘The covariance structure of queues and related processes: a survey of recent work’, Adv. Appl. Probab., 1975, 7, (2), pp. 383–415 (doi: 10.2307/1426082).
-
14)
-
10. Basawa, I., Prabhu, N.: ‘Estimation in single server queues’, Nav. Res. Logist. Q, 1981, 28, (3), pp. 475–487 (doi: 10.1002/nav.3800280311).
-
15)
-
2. Su, Y., Wu, R., Li, R., Duan, W., Luo, J.: ‘A novel measurement-based call admission control algorithm for wireless mobility networks under practical mobility model’. 2010 Int. Conf. on Communications and Mobile Computing, IEEE, 2010, pp. 364–368.
-
16)
-
26. Takacs, L.: ‘Combinatorial methods in the theory of stochastic processes’ (John Wiley and Sons, 1967).
-
17)
-
3. Na, Y., Jun, S., Yuxin, C., Ye, J.: ‘Measurement-based admission control algorithm for wireless IP network’. WRI Int. Conf. on Communications and Mobile Computing, 2009. CMC'09, 2009, vol. 2.
-
18)
-
6. Clarke, A.: ‘Maximum likelihood estimates in a simple queue’, Ann. Math. Stat., 1957, 28, (4), pp. 1036–1040 (doi: 10.1214/aoms/1177706808).
-
19)
-
8. Aigner, D.: ‘Parameter estimation from cross-sectional observations on an elementary queuing system’, Oper. Res., 1974, 22, (2), pp. 422–428 (doi: 10.1287/opre.22.2.422).
-
20)
-
M. Roughan
.
A Comparison of poisson and uniform sampling for active measurements.
IEEE J. Sel. Areas Commun.
,
12 ,
2299 -
2312
-
21)
-
21. Roughan, M.: ‘Fundamental bounds on the accuracy of network perfor- mance measurements’. Proc. 2005 ACM SIGMETRICS Int. Conference on Measurement and Modeling of Computer Systems, 2005, pp. 253–264.
-
22)
-
1. Chen, L., Sun, T., Yang, G., Sanadidi, M., Gerla, M.: ‘Ad hoc probe: path capacity probing in wireless ad hoc networks’. First Int. Conf. on Wireless Internet, 2005. Proc., 2005, pp. 156–163.
-
23)
-
25. Yu, O., Saric, E., Li, A.: ‘Integrated connection-level and packet-level QoS controls over wireless mesh networks’, J. Parallel Distrib. Comput., 2008, 68, (3), pp. 336–347 (doi: 10.1016/j.jpdc.2007.05.012).
-
24)
-
16. Basawa, I., Bhat, U., Zhou, J.: ‘Parameter estimation in queueing systems using partial information’. ., Ohio State University, June 2006.
-
25)
-
26)
-
27. Cox, D., Isham, V.: ‘The virtual waiting-time and related processes’, Adv. Appl. Probab., 1986, 18, (2), pp. 558–573 (doi: 10.2307/1427312).
-
27)
-
17. Chen, T.: ‘Parameter estimation for partially observed queues’, IEEE Trans. Commun., 1994, 42, (9), pp. 2730–2739 (doi: 10.1109/26.317414).
-
28)
-
24. Nicholas, C.L., Duffield, G., Thorup, M.: ‘Optimal combination of sampled measurements, , 2007. .
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-net.2013.0014
Related content
content/journals/10.1049/iet-net.2013.0014
pub_keyword,iet_inspecKeyword,pub_concept
6
6