2000, HCS Research Lab All Rights Reserved
Adaptive Sampling for Network Management
Edwin A. Hernandez, Matthew C. Chidester, and Alan D. George
Highperformance Computing and Simulation (HCS) Research Laboratory
Department of Electrical and Computer Engineering, University of Florida
P.O.Box 116200, Gainesville, FL 326116200
Abstract Highperformance networks require sophisticated management systems to identify sources of bottlenecks
and detect faults. At the same time, the impact of network queries on the latency and bandwidth available to the
applications must be minimized. Adaptive techniques can be used to control and reduce the rate of sampling of
network information, reducing the amount of processed data and lessening the overhead on the network. Two
adaptive sampling methods are proposed in this paper based on linear prediction and fuzzy logic. The performance
of these techniques is compared with conventional sampling methods by c h. I,, ii simulative experiments using
Internet and videoconference ;,,,ni,. patterns. The adaptive techniques are ',..,i li.. ,i Imoreflexible in their ability
to dynamically adjust with fluctuations in network behavior, and in some cases they are able to reduce the sample
count by as much as a factor of two while maintaining the same accuracy as the best conventional sampling
interval. The results illustrate that adaptive sampling provides the potential for better monitoring, control, and
management ofhighperformance networks with higher accuracy, lower overhead, or both.
KEY WORDS: Adaptive sampling; fuzzy logic; linear prediction; network management; SNMP.
1. INTRODUCTION
In network management, accurate measures of network status are needed to aid in planning, troubleshooting, and
monitoring. For example, it may be necessary to monitor the bandwidth consumption of several hundred links in a
distributed system to pinpoint bottlenecks. If the monitoring is too aggressive, it may create artificial bottlenecks.
With too passive a scheme, the network monitor may miss important events. Network query rates must strike a
balance between accurate performance characterization and low bandwidth consumption to avoid changing the
behavior of the network while still providing a clear picture of the behavior. This balance is often achieved through
sampling. Sampling techniques are used to study the behavior of a population of elements based on a representative
subset. In general, the samples are taken periodically at some fixed interval or in some random distribution. Such
sampling reduces bandwidth and storage requirements for the monitored data.
In a highperformance network, sampling overhead must have a minimal impact on the lowlatency application
transactions while high network throughputs require frequent sampling to capture transient behavior. For example,
many distributed applications (e.g. parallel discreteevent simulation, database systems, etc.) may require frequent
synchronization, large data transfers, or both. Network management queries could delay critical data on a single
link and thus reduce the efficiency of the entire application.
Under some traffic loads, simple periodic sampling may be poorly suited to the monitoring task. For example,
during periods of idle activity or low network loads, a long sampling interval provides sufficient accuracy at a
minimal overhead. However, bursts of high activity require shorter sample intervals to accurately measure network
status at the expense of increased sample traffic overhead. To address this issue, adaptive sampling techniques can
be employed to dynamically adjust the sampling interval and optimize accuracy and overhead.
Adaptive sampling monitors network behavior by dynamically adjusting the sampling time interval for each
parameter monitored. When levels of high activity are detected, a shorter sampling interval is employed to measure
the behavior of the network with greater accuracy. When less activity is detected, the sampling interval is
lengthened to reduce sampling overhead. The adaptive algorithms introduced in this paper are used to select the
most representative samples of the population by polling and monitoring key fluctuations in the variables being
measured. Consequently, adaptive sampling techniques allow the management to take place in a lessintrusive
fashion by avoiding unnecessary queries.
This paper introduces two techniques for adaptive sampling. The first technique is based on Linear Prediction
(LP) [1,2]. The LP sampler uses previous samples to estimate or predict a future measurement. The LP logic can be
used in conjunction with a set of rules defining the sampling rate adjustments to make when the prediction is
inaccurate.
The second technique for adaptive sampling makes use of fuzzy logic. Fuzzy logic mimics the decision
processes employed by the human brain [3]. For example, a network administrator may reason that when network
load is low, the sample interval can be increased. The fuzzy logic model relies on previous experiences for defining
the fuzzy set of parameters and boundaries for the fuzzy variables.
In order to gauge the performance of the adaptive techniques, they are compared with systematic sampling. This
technique uses a deterministic interval of time to query the agents. Other nonadaptive sampling techniques include
random sampling where samples are taken at intervals of time determined by a random distribution, and stratified
random sampling where a sample is taken at a random point within a deterministic time interval [4,5].
This paper introduces the concept of adaptive network sampling and provides experimental results with the
various sampling techniques comparing them in terms of accuracy and performance using Internet and
videoconference traffic. In Section 2, the sampling techniques used in this study are defined. Section 3 describes
the experiments and measurement techniques used to compare the performance of the sampling disciplines are
described. The performance results are shown in Section 4. Related research is described in Section 5, while
conclusions and directions for future research are provided in Section 6.
2. SAMPLING TECHNIQUES FOR NETWORK MANAGEMENT
In network management, status information regarding load, latency, queue occupancy, and other parameters is
frequently available in devices such as routers, switches, and network interfaces. Such information is often accessed
through the Simple Network Management Protocol (SNMP) [69]. In SNMP, a Network Management Station
(NMS) queries the network devices, or agents, to periodically assess the status of the network devices or links.
The period of the sampling determines the accuracy of the measured data. Transient activity may not be
accurately detected when the sampling interval is large, while small intervals consume more bandwidth on the
network and require greater storage capacity at the NMS. For example, a burst of high activity lasting only seconds
is likely to go undetected with a sampling interval of several minutes. In an effort to balance accuracy with
sampling overhead, several sampling disciplines have been applied to network managers.
2.1. Conventional Sampling
Traditionally, network management has made use of simple, nonadaptive sampling techniques. Such
techniques use a fixed rule to determine when to sample data in each agent. The sampling rule can be deterministic,
such as in periodic sampling, or it can involve a random component. Introducing randomness has been shown to
improve accuracy in situations where the monitored data is uniform in nature [4]. There are three conventional
methods used by network management systems for sampling of agents:
a) Systematic sampling, or periodic sampling, deterministically samples data at a fixed time interval. Fig.
l(a) shows systematic sampling with a period of T seconds.
b) Random sampling employs a random distribution function to determine when each sample should be
taken. The distribution may be uniform, exponential, Poisson, etc. As shown in Fig. l(b), random
sampling may take a varying number of samples in a given time interval.
c) Stratified random sampling combines the fixedtime interval used in systematic sampling with random
sampling by taking a single sample at a random point during a given time interval. Fig. l(c) shows
stratified random sampling with a time interval of T.
I 7 Y7 77 7 7 Z7 ,7 ;v 7  ; 7i, 77 i,7 i 7 ,
T 2T 3T 4T 5T T 2T 3T 4T 5T T 2T 3T 4T 5T
(a) Systematic sampling (b) Random sampling (c) Stratified random sampling
Fig. 1. Conventional sampling methods.
Since network traffic is frequently periodic, the rate for any of these nonadaptive sampling techniques is
typically set based on the expected average network load, the traffic distribution, or simply to a value that will yield
an acceptably small amount of overhead. If the actual traffic differs from the expected pattern, the measurements
may prove inaccurate or may take an excessive number of samples.
2.2. Adaptive Sampling
Adaptive sampling dynamically adjusts the sampling rate based on the observed sampled data. A key element
in adaptive sampling is the prediction of future behavior based on the observed samples. The two adaptive sampling
techniques presented in this paper differ by the technique used to predict the future behavior of the system. The
Linear Prediction (LP) method attempts to predict the value of the next sample. If the prediction is accurate, the
sampling rate can be reduced. Inaccurate predictions indicate a change in the network behavior and require an
increased sampling rate to determine the new pattern. By contrast, the Fuzzy Logic Controller (FLC) adjusts the
sampling rate based on experiences in past situations with similar sample data. A rule structure is defined that the
FLC can follow to determine the most appropriate action to take given a certain input condition.
2.2.1. Linear Prediction (LP) Method
The LP method proposed for adaptive sampling is based on the Linear Prediction Coefficient (LPC) technique
used by Jacobson and Karels in the congestion control protocol for the Transport Control Protocol (TCP) to predict
the Round Trip Time (RTT) of a packet [10]. LPC employs a lowpass filter to predict future data values. This
technique filters out transient behavior, using the average value of the nontransient data as the prediction for the
next value. This predicted time value was used to adjust the transmission window base on the RTT of previous
packets. The lowpass filter is of the form:
x, =(1a)xM+axR (2.1)
In this equation, x, is the predicted value for the next sample, M is the most recently measured value, and R
represents the average of the previous two samples. The coefficient a ranges between zero and one and is selected
experimentally depending on the network load, where higher values of a reduce the impact of transient values on
the average value.
A similar LPbased technique is used in this paper to control the sampling time. The main difference between
the approach described in this paper and the one used by Jacobson and Karels is that here a variable number of
previous samples are used to calculate the predicted value. Instead of using only the previous two samples, a
window of N samples is used for the prediction. Moreover, the a coefficient for tuning to a specific network load is
no longer employed.
Eq. (2.2) below defines the LP logic for a sampler of order N. As in Eq. (2.1), the value x, represents the
predicted value for the next sample. The vector x holds the value of the previous N samples, where x[N] is the most
recent sample and x[l] is the oldest sample. Sample refers to the value of the current sample. Immediately prior to
taking the next sample, the values in the vector x are shifted such that x[l] is discarded and replaced with x[2], x[2]
is replaced with x[3], and so forth with the value of Sample replacing x[N]. Eq (2.2) is designed to work with non
decreasing SNMP byte counters, and therefore the difference between any two values in x will always be non
negative. A second vector, t, records the time that each sample is taken and is shifted in the same manner as x, with
the time at which Sample was taken replacing t[N]. Since the period between samples is not necessarily constant,
the next value is predicted based on the average rate of change in the previous N samples.
S= N+ T x[i +1] x[i]+ Turren x x] (22
N1 t[i + 1] t[i] N1 (t[N]t[1]
The prediction is then used as shown in Fig. 2. The predicted output, xp, which has been derived from the
previous N samples, is then compared with the actual value of Sample. A set of rules is applied to adjust the current
sampling interval, ATcurrent = t[N] t[N1], to a new value, ATNext, which is used to schedule the next management
query. The rules used to adjust the sampling interval compare the rate of change in the predicted sample value, x,
x[N], to the actual rate of change, Sample x[N]. This rate, m, is given by Eq. (2.3).
ATCurrent
ATAT
Next
Rules 
Sample _ vector x LP logic
(N samples) (order N)
Fig. 2. LPbased adaptive sampling.
x, x[N]
m = P (2.3)
Sample x[N]
The rate of change provided by Eq. (2.3) will take on a value near unity when the predicted behavior is close to
the actual behavior. The range of values that satisfy this condition is defined as mmm < 1 < mm,. If m is below mm,,,
the measured parameter is changing faster than the prediction. Such behavior indicates more activity than predicted,
so the sampling interval should be decreased to yield more accurate data values on which to base future predictions.
Conversely, if m is above mmax, the measured value is changing more slowly than the prediction so the sampling
interval can be increased. Note that m can be undefined if the current value of Sample is unchanged from that of the
previous sample, x[N]. Since this condition is indicative of an idle network, the sampling interval is increased
exponentially.
Table I. LP rules for adjusting sample interval.
Calculated m Value Next Sample Interval
m < mmn ATNext = m X ATurrent
mmn < m < mmax TNext ATCurrent
mma < m ATt = ATcurrent + 1 sec
m undefined ATNt = 2 x ATcurrent
Table I lists the specific rules for generating AText given the current value of m. Based on the results of
experiments described later, mmm and mmax are set to 0.9 and 1.1 respectively. These values were selected because
they provided good performance over a range of traffic types. An additional constraint can be used to limit the
range of possible values for ATet. For example, in this paper, the sampling interval is restricted to be between 1
and 10 seconds regardless of the output of the LP logic. The lower bound is applied to limit the sampling to a rate
the manager and agents can realistically service, while the upper bound ensures there is some minimum set of
samples on which to base future predictions.
2.2.2. Fuzzy Logic Controller (FLC) Method
Rather than attempting to predict the exact values of future samples, fuzzy logic applies "rules of thumb" to the
observed behavior to adjust the future sampling rate in a reasonable fashion. Fuzzy logic has shown promising
results in nonlinear systems, especially those that are difficult to model or where an exact model is impossible. The
FLC mimics human reasoning by applying a set of rules based on one or more premises and a single implication.
For example, if the premises "network load is very high" and "the sampling interval is somewhat low" are found to
be true, the implication "reduce the sampling interval by a medium amount" might be taken. In this section, an FLC
is used to adaptively control the sampling rate of the network monitor.
The proposed method for adaptive sampling using fuzzy logic is shown in Fig. 3. The inputs of the FLC are the
current sampling interval, ATcurrent, and the difference between the last two sample values, AX. The FLC output,
Fo,, corresponds to the amount of time to increment or decrement the current sampling interval, ATcurrent. This
adjustment yields a new value for the next sampling interval, ATxet. The following subsections describe the FLC,
including the definition of the membership functions, the fuzzy rules, and the defuzzification process.
ATN,
Fig. 3. Adaptive sampling based on fuzzy logic.
2.2.3. FLC Membership Functions
Unlike traditional digital systems, fuzzy logic differentiates between several levels such as "high," "medium,"
and lo\\" when making decisions. The numerical ranges assigned to each of the levels are defined in a membership
function. The FLC makes use of a set of membership functions such as those illustrated in Fig. 4 to determine a
suitable output value given the state of the input values. In this case, there are two membership functions for the
inputs and one for the output. For the two input membership functions, AX and ATcurrent, the horizontal axis
corresponds to the value of the input. A given value of the input is interpreted as being in one or more fuzzy states.
For the AX input, the fuzzy states are No( I1,,i.i,. (NC), ( hIo, i.\l/Jit (CS), ( i.,, i.. Low (CL), C /i,, ,.. ,...,lI1
(CM), and (C/'. .. High (CH). The ATcurrent input falls into the categories of Small (S), SmallMedium (SM),
Medium (M), MediumLarge (ML), and Large (L).
The output of the FLC is also defined in terms of fuzzy variables. In Fig. 4, the output of the FLC, Fo,,, falls
into one of five fuzzy states: DecreaseHigh (DH), DecreaseLow (DL), No( /h,, . (NC), IncreaseLow (IL), and
IncreaseHigh (IH).
NC CS CL CM CH S SM M ML L DH DL NC IL IH
1 1 1
I I I
oo
0 P/6 P/3 P/2 2P/3 5P/6 P 0 M/6 M/3 M/2 2M/3 5M/6 M L 2L/3 L/3 0 L/3 2L/3 L
(a) AX (b) ATrrent (c) Fo,
Fig. 4. Membership functions used by the FLC.
In Fig. 4, the labels at the top of each peak in the graph represent the fuzzy states for the given input or output
variable. The yaxis shows the fractional membership in each state for a given value along the xaxis. The xaxis
value corresponds to the numerical value of the input or output variable. The membership function for input AX
describes the amount of change in measured throughput (in terms of SNMP bytecount) between successive samples
and ranges from 0 to P bytes. Similarly, the membership function for input ATcurre,,t denotes the current sampling
interval and ranges from 0 to K seconds. Finally, the membership function for output F,, indicates the amount of
increase or decrease to be applied to the sampling interval and ranges from L to +L seconds.
Table II. Parameter values selected for the membership functions in the FLC.
Type of Traffic P (bytes) K (sec) L (sec)
Internet 0.4 x Ste,, 12 2.0
Videoconference 4.0 x S,, 12 0.5
The appropriate selection of the membership parameters P, K, and L requires some understanding of the traffic
behavior. When determining the values for these FLC parameters, it is desirable to select values that lead to a
relatively uniform spread in frequency of occurrence of inputs across the states of the membership function. For
instance, with parameter P, the goal is select the value that distributes the input data in a relatively uniform fashion
across all five of the stages, from NC to CH. Based on several tuning experiments with the Internet and
videoconference traffic models described in the next section, the values selected for this controller were determined
as shown in Table II. In this table, the variables SInern and Sv,deo refer to the peak throughputs (i.e. the maximum
change in SNMP transmitted bytecount measured over a span of one second) in the Internet and videoconference
data, respectively. These values are multiplied by a scaling factor to produce the desired value for parameter P. For
Internet traffic, a small scaling factor is selected, which allows the FLC to respond to smaller changes in network
behavior. Coupled with a larger value of L, the FLC can make large adjustments in the sampling interval in
response to small changes in network behavior. The videoconference traffic employs a larger scaling factor and
smaller value of L since the periodic nature of the traffic requires less drastic adjustments to sample rate. Although
beyond the scope of this paper, the use of an FLC in a network with arbitrary traffic would require a training period
during which the optimal spread of these three parameters across the fuzzy states in the membership functions could
be ascertained.
2.2.4. FLCRules
In addition to the membership functions for the inputs, the FLC needs a fuzzy set of rules to map the input
values to an output response. The twentyfive statements shown in Table III represent a proposed set of fuzzy rules
for the FLC. Each row in this table provides two premises along with a single implication. In the table, the first and
second columns are correlated using the logic operator AND, which is analogous to the intersection in set theory.
For example, the fuzzy logic expression shown in the first row of the table indicates that given no measured changes
in the input, AX, and a small sampling interval, ATcrrent, then Fo, should cause the sampling interval to be increased
by a high amount.
Table III. Rules for the fuzzy controller.
Rule AX ATcurrent Fo,_
1 NoChange (NC) Small (S) IncreaseHigh (IH)
2 NoChange (NC) SmallMedium (SM) IncreaseHigh (IH)
3 NoChange (NC) Medium (M) IncreaseLow (IL)
4 NoChange (NC) MediumLarge (ML) IncreaseLow (IL)
5 NoChange (NC) Large (L) NoChange (C)
6 ChangeSlight (CS) Small (S) IncreaseHigh (IH)
7 ChangeSlight (CS) SmalllMedium (SM) IncreaseLow (IL)
8 ChangeSlight (CS) Aledium (M) NoChange (NC)
9 ChangeSlight (CS) AlediumLarge (ML) DecreaseLow (DL)
10 ChangeSlight (CS) Large (L) DecreaseLow (DL)
11 ChangeLow (CL) Small (S) IncreaseLow (IL)
12 ChangeLow (CL) SmallMedium (S/M) NoChange (NC)
13 ChangeLow (CL) Medium (M) DecreaseLow (DL)
14 ChangeLow (CL) MediumLarge (ML) DecreaseHigh (DH)
15 ChangeLow (CL) Large (L) NoChange (AC)
16 ChangeMedium (CM) Small (S) DecreaseLow (DL)
17 ChangeMedium (CM) SmallMedium (SM) DecreaseLow (DL)
18 ChangeMedium (CM1) Medium (M) DecreaseHigh (DH)
19 ChangeMedium (CM) MediumLarge (ML) DecreaseHigh (DH)
20 ChangeMedium (CM) Large (L) DecreaseHigh (DH)
21 ChangeHigh (CH) Small (S) DecreaseLow (DL)
22 ChangeHigh (CH) SmallMedium (SM) DecreaseHigh (DH)
23 ChangeHigh (CH) Medium (M/) DecreaseHigh (DH)
24 ChangeHigh (CH) MediumLarge (ML) DecreaseHigh (DH)
25 ChangeHigh (CH) Large (L) DecreaseHigh (DH)
The "fuzziness" of the FLC stems from the fact that a given input value may correspond to multiple fuzzy
states. For example, referring to Fig. 4(a), if AX has a value between P/3 and P/2 then it would be a partial member
in the CS state and a partial member in the CL state. Such fuzziness makes it hard to determine a fixed expression
for the output value from Table III. In order to find the nonfuzzy output, the defuzzification process takes place.
This process is composed of an inference method that makes use of the rules and membership functions to produce
the control output. The correlationproduct method is used here for defuzzification with the FLC [3].
2.2.5. FLC Defuzzification Process
Consider the example in Fig. 5 using the correlationproduct method for defuzzification. The figure
assumes the input values of AXo and ATo for AX and ATcurre,,,, respectively. The membership functions indicate two
degrees of membership for each input: a value of a in the CS state and P in the NC state for the input AXo, and a
value of 8 in the ML state and e in the M state for To. There are three steps in the defuzzification process using the
correlation product method. The first step is to find the set of all rules that correspond to the input membership
states by considering all permutations of the states. In this example, the rules 3, 4, 8, and 9 from Table III apply.
The second step is to choose the minimum value of the degree of membership of the inputs among the matching
rules. For instance, rule 3 involves the fuzzy states NC and M in the membership functions AX and ATcurret,
respectively. The degree of membership for AXo is P and for ATo is e. Since e < P, then e is the minimum input
value for rule 3. The third step scales the shape of the membership function for the Fo,, value of each rule by the
minimum input value determined in the previous step. In the example below, rule 3 defines the IL state for Fo,,, so
the IL shape is scaled down by e. This process is repeated for all the rule graphs and the resulting scaled shapes are
combined. The center of mass of the combined graph is calculated using Eq. (2.4) [3,11]. The resulting outcome, z,
corresponds to the numerical value used for Fo,.
AX
NC CS CL CM CH
a 
83
0 P/6 P/3 P/2 2P/3 5P/6 P
AX
AT
Current
0 M/6 M/3 M/2 2M/3 5M/6 M
AT
Fouto
L 2L/3 L/3 0 L/3 2L3 L
rule #3
IL
0 L/3 2L/3
out
rule #4
IL

rule #8
NC
L/3 0 L/3
rule #9
DL
0 L/3 2L/3 2L/3 L/3 0
Fo1 Union of matching inference rules and resulting outcome
(center of mass defuzzification)
DL NC IL
a L Ux 0T L+V L + Lo L
j a+(0)E+P+e
z= 3 3 3
n \a+E+P+E
2L/3 L/3 0 L/3 2L3
Fig. 5. Correlationproduct inference method and defuzzification process.
'u, Xv,
z= =1 (2.4)
J=1
For each of the n scaled shapes in the combined graph, the center of mass equation uses Uj and V, to represent
the values for the peak of the shape on the xaxis and yaxis, respectively. For example, Fig. 5 shows the
combination of the scaled rule graphs where the Uj values are L/3, L/3, 0, and L/3, and the V, values are e, fP, e, and
a, from rules 3, 4, 8, and 9, respectively. The result of this equation represents the numerical (i.e. nonfuzzy) output
for membership function Fou,. Future sampling queries are rescheduled based upon this scalar result as shown in
Fig. 3.
3. EXPERIMENTS AND MEASUREMENT TECHNIQUES
This section describes a set of simulative experiments to compare the adaptive and systematic sampling
techniques for network management. The computing testbed that generated the traffic in the simulation is described
in this section, as well as the metric used to compare the sampled data sets.
3.1. Traffic Models and Sampler Simulation
Two traffic traces of network activity are used as data models for the experiments. The two were selected to
represent the opposite ends of the spectrum of potential traffic that might be encountered in a highperformance
network, one characterized by bursty, periodic traffic and the other by streaming, periodic traffic.
The first trace is a twohour sample taken from the outgoing 100 Mb/s Fast Ethernet port on a Fore Systems
PowerHub 7000 router providing Internet connectivity to a laboratory with dozens of computers during a busy part
of the day. This Internet trace is representative of random, bursty TCP/IP network data and includes traffic from
Hypertext Markup Language (HTML), File Transfer Protocol (FTP), and Telnet activity. This type of data should
be ideally suited to adaptive sampling because the sample rate can be adjusted to compensate for the level of loading
on the network at any given time. Thus, the goal with adaptive sampling for this type of traffic is to reduce the
sample count while maintaining the accuracy achieved by the equivalent systematic sampler. Or, conversely, the
goal is to increase the accuracy while maintaining the same sample count as the equivalent systematic sampler.
A second trace was captured from a port of a Fore Systems ASX200BX Asynchronous Transfer Mode (ATM)
switch that was connected directly to a Fore Systems AVA300 ATM audio/video decoder. This connection hosts a
continuously running videoconference feed transferring variablebitrate compressed video and uncompressed audio
at 30 frames/second over a dedicated 155 Mb/s ATM link. The resulting videoconference trace is representative of a
highly periodic pattern of network traffic. The periodic nature of this data may not be better suited to adaptive
sampling techniques than the systematic sampler. However, the adaptive sampling algorithm can dynamically select
a sampling interval that accurately measures the data. By contrast, a systematic technique requires the proper
interval to be correctly selected a priori by the network administrator, which is not easily realized.
Each trace was taken by measuring the cumulative number of bytes received over a twohour period using
SNMP queries. The resolution of the measurements was 0.1 seconds, yielding a total of 72,000 samples. The
adaptive and systematic sampling methods were then simulated and evaluated by resampling the data in each trace.
For example, a systematic sampling with T = 1 sec would select every tenth element from the original data, yielding
a new trace with 7,200 samples. During the simulation, all sampling intervals were rounded to the nearest 0.1
seconds so that they correspond to a single measurement from the original trace. For simplicity, the throughput
measurements gathered by the systematic and adaptive samplers with these traces are all normalized to the peak
value stored in the baseline data model (i.e. either Internet or videoconference) used in the experiment.
3.2. Metric for Comparison
In order to compare the performance of the adaptive sampling techniques with the systematic baseline, a
measure of accuracy is needed. The sum of squared error metric [2] for comparing two Nsample sets, as shown in
Eq. (3.1), is used in Section 4.3 to compare the accuracy of the different techniques. This expression makes a point
bypoint comparison between the reference and sampled signals using the normalized magnitude of the
instantaneous throughput. The reference signal, fR(n), is a systematic sampling of the original trace of Internet or
videoconference data with a period of T, = 1 sec. The comparison signal, fc(n), is the output from one of the
adaptive or systematic sampling methods applied to the original trace. Since Eq. (3.1) requires sets with an equal
number of samples, both the reference and the comparison traces are resampled to 72,000 samples. Linear
interpolation is used to produce the added points.
N
Error = f (f n) f (n))2 (3.1)
n=1
4. RESULTS AND ANALYSES
This section presents results and analyses from experiments conducted to ascertain the performance of the
systematic and adaptive sampling algorithms as applied to the Internet and videoconference traces. In the next two
subsections, a qualitative comparison of the signals produced by several systematic and adaptive sampling
measurements is provided. Then, a quantitative assessment of the adaptive sampling techniques is conducted by
comparing the adaptive techniques to an equivalent systematic baseline in terms of sample count and meansquared
error relative to the reference trace.
4.1. Systematic Sampling Measurements
In this section, the effect of sampling rate is qualitatively illustrated using traditional, systematic sampling
techniques. Fig. 6 and Fig. 7 show the normalized throughput sampled using fixed sampling intervals (T,) from one
to ten seconds on the Internet and videoconference traces, respectively. The shapes for Internet and videoconference
traffic at T, = 1 sec shown in Fig. 6(a) and Fig. 7(a) are used as the reference value in the comparisons of the
different sampling disciplines.
Both figures demonstrate a "smoothing" effect at larger sampling intervals. This effect results from the fact
that the samples actually measure a cumulative SNMP bytecount rather than a throughput. Therefore, at higher
values of Ts, the measured bytecount is averaged over a longer interval of time. The result is that large throughput
spikes are averaged with periods of low activity. This effect both reduces the magnitude of the throughput spikes as
well as filters much of the variation from the actual traffic pattern.
1
04   
1 0 2     
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(a) T,= 1 sec (7200 samples)
08
0 6       
S06
S04   
02
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
1
08 il U 
02
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(b) T,= 2 sec (3600 samples)
1
08
o 06           
04  
1 02 1  
0
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(c) T, = 5 sec (1440 samples) (d) T,= 10 sec (720 samples)
Fig. 6. Measured traffic patterns of Internet data using systematic sampling.
08
06
02
0
0 
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(a) T,= 1 sec (7200 samples)
06  
04        
S02
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
1
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(b) T,= 2 sec (3600 samples)
06
o04
o
E 02
0
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
0 4     
0     
Time (Seconds)
(c) T, = 5 sec (1440 samples) (d) T,= 10 sec (720 samples)
Fig. 7. Measured traffic patterns ofvideoconference data using systematic sampling.
The measurements conducted at T, = 10 sec for the Internet data in Fig. 6 indicate a maximum throughput of
only about 55% of the peak value at the reference. Fig. 7 illustrates a similar situation with videoconference traffic.
For example, in the trace sampled at T, = 5, the highly fluctuating behavior found at the reference is completely lost.
In fact, the variation observed in Fig. 7(c) is only approximately 25%, whereas the reference indicates a variation of
5070% of the maximum peak throughput.
Since highperformance management services require precise throughput estimations, an inappropriate selection
of the sampling time would lead to inaccurate observations. The following section will show how adaptive
sampling can be used to maintain accurate measurements for Internet and videoconference traffic.
4.2. Adaptive Sampling Measurements
The results of applying the adaptive sampling techniques to the Internet trace using second, third, and fourth
order LP samplers and the FLC sampler are shown in Fig. 8. The secondorder LP sampler was able to reduce the
sample count to about 51% of the reference with almost no discernable change in the peak magnitude or variation of
throughput behavior. As the order of the LP sampler is increased, the number of samples are further reduced with
I
only a small visible loss in peak magnitude and variation. The FLC approach yields the fewest number of samples
at 29% of the reference. The FLC shows more variation than the higherorder LP samplers but with marginally
lower peaks.
Fig. 9 shows the results of applying the adaptive sampling techniques to the periodic videoconference trace.
The LP techniques reduce the number of samples to approximately 65% of the reference but at a slight reduction in
signal variation. The FLC achieves shows slightly less variation at 43% of the reference sample count. However,
the accuracy of the adaptive sampling techniques on the videoconference data is not nearly as good as it is with the
Internet data.
Although a qualitative comparison of the adaptive sampling techniques shows the ability of the adaptive
sampling techniques to reduce sample count while retaining a certain degree of accuracy, an accurate picture of the
performance is difficult to ascertain. For example, a simple systematic sampler with T, = 2 sec is also able to reduce
the number of samples by 50% while still retaining much of the characteristics in the reference signal. Therefore, it
is necessary to quantify the accuracy of the adaptive and systematic sampling techniques.
001 , .. .., I  I 
1
08   
S 06 
04 
02
0
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(a) Secondorder LP (3728 samples)
08
oo i L . ..
S  
04
08
= 06
O 04
o 02
0 2
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(b) Thirdorder LP (3028 samples)
0
z 0
0
0
1
6      
6
4
2
0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds) Time (Seconds)
(c) Fourthorder LP (2641 samples) (d) FLC (2095 samples)
Fig. 8. Measured traffic patterns of Internet data using adaptive sampling.
U
08
E
04
0z02
0
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(a) Secondorder LP (4769 samples)
08
2 06
04   
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(c) Fourthorder LP (4621 samples)
08
06
04
o 02
0
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(b) Thirdorder LP (4651 samples)
08
2 06
04
z 02
02
0 1000 2000 3000 4000 5000 6000 7000
Time (Seconds)
(d) FLC (3095 samples)
Fig. 9. Measured traffic patterns ofvideoconference data using adaptive sampling.
4.3. Quantitative Analysis of Adaptive Samplers
It is difficult to quantitatively compare the performance of the adaptive sampling techniques to an equivalent
systematic approach when the measurements differ in both the number of samples and the squarederror relative to
the reference. In general, traces with more samples show a lower error. Therefore, two forms of experiments were
conducted to compare the adaptive and systematic sampling methods, first with a fixed number of samples and then
with a fixed error. In both cases, a single run of the adaptive sampling techniques was used. As demonstrated in the
previous section, each technique yielded a different number of samples with a different relative error when
compared to the reference. In the fixed samplecount comparison, a systematic sampling interval was selected for
each adaptive sampling result such that it would contain the same number of samples. The relative error from each
adaptive sampler was then calculated and compared to its equivalent systematic sampler. For the constanterror
comparison, a chart of sample count versus relative error was made for a range of systematic sample intervals.
Using this chart, the minimum number of systematically distributed samples necessary to achieve a certain relative
error was determined.
5500
5000  U Adaptive Sampler U Systematic Sampler]
S4500
4000         
3000   
2500
S2000
1500
1000
500
Secondorder Thirdorder Fourthorder FLC
LP LP LP
Adaptive techniques
700
S Adaptive Sampler U Systematic Sampler
500
400           
LE 300
100   
0
Secondorder Thirdorder Fourthorder FLC
LP LP LP
Adaptive techniques
(a) Sample count with constant error (b) Error with constant sample count
Fig. 10. Sample count and error variation with Internet traffic.
Fig. 10 shows the results of the experiments conducted with Internet traffic. In Fig. 10(a), the amount of
sampling error is held constant and then the sample count is measured for each of the adaptive samplers and
compared to a systematic sampler with the same number of samples. In Fig. 10(b), the sample count is held constant
and the error is measured for each case. As the results indicate, each of the four adaptive samplers outperforms its
corresponding systematic sampler. The greatest improvement is seen with the FLC sampler. The FLC sampler
achieves nearly twice the performance of its equivalent systematic sampler, which can be interpreted either as
decreasing the sample count by a factor of two for a given level of accuracy or as increasing the accuracy by a factor
of two for a given number of samples. In contrast, the secondorder LP sampler is only marginally better in
performance than its systematic counterpart. As previously hypothesized, traffic of a bursty, periodic nature is
found to be well suited for the capabilities of adaptive sampling, since these samplers are able to dynamically adjust
their sampling rate with periods of increased and decreased activity on the network.
rtive Sampler U Systematic Sampler 
700 
S0Adaptive Sampler U Systematic Sampler
600 
500 
400           
S300      
S300 
10020  
1 
Secondorder Thirdorder Fourthorder FLC Secondorder Thirdorder Fourthorder FLC
LP LP LP LP LP LP
Adaptive techniques Adaptive techniques
(a) Sample count with constant error (b) Error with constant sample count
Fig. 11. Sample count and error variation with videoconference traffic.
7 [i
5500
5000 
U 4500
. 4000
E 3500
a 3000
S2500
, 2000
E1500
Z 1000
500
I
u
Fig. 11 shows the results of the same experiments with the videoconference traffic. As before, Fig. 11(a)
displays the measurements of sample count for a fixed level of accuracy and Fig. ll(b) displays the measurements
of accuracy for a fixed number of samples. The results in these experiments with videoconference traffic are
markedly different from those with the Internet traffic. The LP samplers show slightly higher error or slightly larger
sample counts than their systematic counterparts. The FLC sampler achieves slightly lower sample counts and
slightly smaller error rates. Given the nature of the traffic being sampled, these results are also promising. When
dealing with traffic of a periodic nature, as is the case with the videoconference, systematic sampling is effective
provided that the appropriate sampling interval can be identified and employed. Thus, the primary goal of using an
adaptive sampler for periodic traffic is to match the performance of the best systematic sampling rate. This
performance is achieved and even surpassed in results from the experiments with videoconference data by using the
FLC sampler, and the LP samplers show only a marginal degradation in performance. An advantage of adaptive
sampling is that this optimal rate can be dynamically adjusted if the frequency or type of periodic traffic changes.
Taken together, the results of the experiments with Internet and videoconference traffic confirm that adaptive
sampling can play an important role in decreasing the load on the network and the network manager, increasing the
accuracy of the measurements, or both. While these two traffic models do not of course encompass the entire
universe of possible traffic patterns, they are reasonably representative of the boundaries of that universe in terms of
bursty, periodic traffic versus streaming, periodic traffic. Thus, these results illustrate that adaptive sampling
provides the potential for better monitoring, control, and management of highperformance networks.
5. RELATED WORK
Several researchers have studied sampling techniques in networks. Claffy et al. studied three nonadaptive
sampling methodologies: systematic (i.e. periodic), random, and stratified random sampling [4]. Their results
indicated that stratified random sampling has better accuracy when the three methods are used to capture the same
number of samples. Cozzani and Giordano made use of conventional sampling methods and studied their effects in
quality of service measurements for ATM networks [5]. Drobisz proposed a network capture device for Gigabit
Ethernet, adapting CPU utilization based on traffic burst anticipation by increasing CPU usage when a bursty period
of traffic was expected [12].
Fuzzy and LP adaptations have been used in network applications for congestion, admission, and flow control.
Jacobson and Karels studied flowcontrol protocols based on filters for the congestion control algorithm for TCP
[10]. Kalampoukas et al. applied filterbased techniques for window adaptation in TCP [13]. These methods
control protocol parameters based upon network traffic behavior. The adaptive sampling techniques presented
herein employ a similar approach, but by contrast the purpose is for more efficiency in sampling of network
management data with the goal of reducing sample count and/or increasing accuracy in a dynamic traffic
environment.
Fuzzylogic controllers have also been studied for ATM networks. Bonde and Ghosh [14] and Catania et al. [15]
proposed queue managing and congestion control based on fuzzy sets. Cheng and Chang suggested a fuzzy
architecture for both congestion and call admission control in ATM networks [11], and discovered that congestion
control, using fuzzy logic, performed better than the leaky bucket algorithm inherent to ATM. The FLCbased
sampling technique proposed herein uses an adaptation of the methods applied in these studies on control of ATM
networks. Whereas these other studies used fuzzy logic to control parameters in the ATM network, our FLCbased
adaptive sampling applies fuzzy logic to the monitoring of network traffic in any environment. Givan and Chong
applied Markov and Bernoulli processes to model network traffic behavior and to create learning algorithms to
predict future traffic measurements in a switchbased application [16]. By contrast, the adaptive techniques
presented herein assume minimal knowledge of the traffic model, instead applying linear prediction and fuzzylogic
to adapt to any traffic pattern.
6. CONCLUSIONS
This paper has presented two techniques to adaptively monitor network behavior. One approach is based on
using linear prediction to dynamically alter the sample rate based on the accuracy of the predictions, where
inaccurate predictions indicate a change in the network's behavior and result in a smaller sampling interval. The
second approach models the cognitive process of a human network manager by using fuzzy logic. When certain
predetermined conditions are met, such as an increase in network traffic, corresponding actions are taken such as a
decrease in sampling interval.
These adaptive techniques are shown to perform well on random, bursty data such as conventional Internet
traffic. All approaches are able to reduce the sample count while maintaining the same degree of accuracy as the
best systematic sampling interval. Equivalently, all approaches are able to increase accuracy while maintaining the
same sample count. The higherorder LP samplers perform better than the lowerorder ones, while the FLC sampler
shows the greatest reduction in sample count. The reduced sample count is an important factor for network
management in highperformance networks. Accurate measurements are required to find bottlenecks, while at the
same time the impact on the network must be minimized to allow applications to take advantage of the low latency
and high throughputs such networks provide.
For periodic data such as that found in a videoconference environment, the adaptive sampling approaches
perform comparably to the best systematic approach in terms of accuracy and sample count. In particular,
systematic sampling marginally outperforms the LP samplers while the FLC approach shows a slight improvement
over systematic sampling. However, perhaps more importantly, the adaptive techniques have the ability to adjust to
changing traffic loads, and therefore in an environment with dynamic traffic patterns, such as with data
communications, multimedia and integrated services, adaptive sampling techniques hold the potential to outperform
systematic sampling with its preselected and fixed sampling interval.
In general, the evidence suggests that the fuzzylogic adaptive technique provides more flexibility and better
performance than the LP methods. The main disadvantages of the FLC are the selection of the boundaries of the
membership functions and the computational overhead required to implement it. Future work is needed to study
methods to optimally tune the parameters of the adaptive samplers. One possible area of concentration is the
development of an autonomous adaptive manager using fuzzy logic that dynamically adapts the parameters for its
membership functions while the monitoring process is underway. Furthermore, the FLC could maintain a history of
past samples as in the LP approach, at the cost of even more storage and computational overhead, but could
potentially yield better results. In addition, other approaches for the design of adaptive samplers and management
systems are worthy of investigation, such as samplers based on neuralnetwork or neurofuzzy controllers.
ACKNOWLEDGEMENTS
This research was sponsored in part by the National Security Agency. A portion of this work was made possible by
a Fulbright Foreign Graduate Student Fellowship from the Institute of International Education (IIE).
REFERENCES
1. S. Kay, Modern Spectral Estimation Theory and Application, Prentice Hall, USA, 1988.
2. B. Widrow and S. Stears, Adaptive signalprocessing, Prentice Hall, USA, 1985.
3. E. Cox, The Fuzzy Systems Handbook, Second Edition, AP Professional, USA, 1999.
4. K. Claffy, H. Braun, and G. Polyzos, Application of Sampling Methodologies to Network Traffic
Characterization, Proceedings ofACM SIGCOAII '93, San Francisco, California, pp. 194203, September
1993.
5. I. Cozzani and S. Giordano, Traffic Sampling Methods for Endtoend QoS Evaluation in Large
Heterogeneous Networks, Computer Networks and ISDNSystems, Vol. 30, No. 1618, pp. 16971706, 1998.
6. J. Case, M. Fedor, M. Schoffstall, and J. Davin, A Simple Network Management Protocol, RFC1157, SNMP
Research, May 1990.
7. M. Rose, The Simple Book: An Introduction to Internet Management, Second Edition, Prentice Hall, 1994.
8. S. Feit, N li' A Guide to Network Management, McGraw Hill, USA, 1995.
9. W. Stallings, N 1i, ', \ i 1'2, and CMIP. The Practical Guide to Network Management Standards, Addison
Wesley, Reading, Massachusetts, USA, 1993.
10. V. Jacobson and M. Karels, Congestion Avoidance and Control, Proceedings of the ACM SIGCOAMM '88,
Stanford, California, pp. 273288, August 1988.
11. R. Cheng and C. Chang, Design of a Fuzzy Traffic Controller for ATM Networks, IEEE/ACMA Transactions on
Networking, Vol. 4, No. 3, pp. 460469, June 1996.
12. J. Drobisz, Adaptive Sampling Methods for High Speed Networks to Determine Traffic Statistics including the
Hurst Parameter, Master's Thesis, University of South Florida, 1998.
13. L. Kalampoukas, A. Varma, and K. Ramakrishman, Explicit Window Adaptation: A Method to Enhance TCP
performance, Technical Report, University of California Santa Cruz, UCSCCRL9721, August 1991.
14. A. Bonde and S. Ghosh, A Comparative Study of Fuzzy versus 'Fixed' Threshold for Robust Queue
Management in CellSwitching Networks, IEEE/ACM Transactions on Networking, Vol. 2, No. 4, pp. 337
344, August 1994.
15. V. Catania, G. Ficili, S. Palazzo, and D. Panno, A Comparative Analysis of Fuzzy versus Conventional
Policing Mechanisms for ATM networks, IEEE/ACM Transactions on Networking, Vol. 4, No. 3, pp. 449459,
June 1996.
16. R. Givan and E. Chong, Intelligent and Adaptive Management of Multiclass Networks, Technical Report,
Purdue University, ARPA No. G371, April 1999.
