On the performance of a simple packet rate estimator

Abstract

Abstract-This paper proposes a simple packet rate estimator that can be very useful in predicting the rate of network traffic. The quality and performance of the estimator is evaluated and compared with three popular rate estimators that were originally designed for estimating bit rate. The proposed estimator is highly cost effective as its computation is not carried out upon the arrival of each incoming packet. In addition, the computation is simple and does not depend on the measurement of interarrival times of packets. We evaluate and compare the quality and performance in terms of agility, stability, accuracy and cost. The performance evaluation is conducted using discrete-event simulation that produces synthesized bursty traffic with empirical packet sizes.

An Analytical Model for Estimating Cloud Resources of Elastic Services

“…The mean departure rate of all instances is denoted by c, which also represents the overall throughput of the system. For estimating the average workload k, the reader can refer to prior work on estimating an instantaneous dynamic workload published in [23][24][25][26].…”

Implementation and experimental evaluation of a simple packet rate estimator

“…In previous work [15], we briefly presented and evaluated the performance of a total of four different packet arrival rate estimators. One of these estimators is a novel and simple estimator that we proposed.…”

A model of optimal burst assembly for delay reduction at ingress OBS nodes

“…The BCP is sent only when the timer reaches time t 1 ; the estimation time is thus equal to T a -T o . The estimated length ( L e ) is then calculated by using our improvement of the TW-EWMA algorithm [13], which is presented in the next section.…”

“…The original TW-EWMA algorithm [13] estimates the rate of packets arriving in a time window instead of all arriving packets. It is significant for OBS networks in terms of computation costs, due to the huge amount of packets arriving at ingress OBS nodes.…”

“…In [13], α was fixed at 0.3. This fixed value has an obvious negative impact on the estimated results, when the burstiness of traffic is popular in real networks.…”