The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Este artigo propõe um novo modelo que pode aproximar a distribuição do tempo de atraso na Internet. É bem sabido que o tempo de atraso nos links de comunicação segue a distribuição exponencial. Contudo, os modelos anteriores não conseguem explicar a distribuição quando um link de comunicação está fortemente sobrecarregado. Este artigo propõe usar o M/M/S(m) modelo para a Internet. Aplicamos nosso modelo aos resultados da medição. Este artigo trata do atraso unidirecional porque reflete as características reais dos links de comunicação. A maioria das estatísticas de medição na Internet tem sido baseada no atraso de ida e volta entre dois nós finais. Essas características são facilmente medidas enviando pacotes de amostra de um nó para outro. O lado receptor ecoa os pacotes. No entanto, os resultados nem sempre são úteis. Um link de comunicação de longa distância, como uma linha alugada, possui duas fibras ou fios diferentes para cada direção: um link de entrada e um link de saída. Quando o link está sobrecarregado, o tráfego em cada link é bem diferente. A medição do atraso unidirecional é especialmente importante para comunicações multimídia, porque as transmissões de áudio e vídeo são essencialmente tráfego unidirecional.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Jun-ya KATO, Atsuo SHIMIZU, Shigeki GOTO, "End-to-End Delay Distribution on the Internet" in IEICE TRANSACTIONS on Information,
vol. E82-D, no. 4, pp. 762-768, April 1999, doi: .
Abstract: This paper proposes a new model which can approximate the delay time distribution in the Internet. It is well known that the delay time in communication links follows the exponential distribution. However, the earlier models cannot explain the distribution when a communication link is heavily overloaded. This paper proposes to use the M/M/S(m) model for the Internet. We have applied our model to the measurement results. This paper deals with one-way delay because it reflects the actual characteristics of communication links. Most measurement statistics in the Internet have been based on round-trip time delay between two end nodes. These characteristics are easily measured by sending sample packets from one node to the other. The receiver side echoes back the packets. However, the results are not always useful. A long distance communication link, such as a leased line, has two different fibers or wires for each direction: an incoming link, and an outgoing link. When the link is overloaded, the traffic in each link is quite different. The measurement of one-way delay is especially important for multimedia communications, because audio and video transmissions are essentially one-way traffic.
URL: https://global.ieice.org/en_transactions/information/10.1587/e82-d_4_762/_p
Copiar
@ARTICLE{e82-d_4_762,
author={Jun-ya KATO, Atsuo SHIMIZU, Shigeki GOTO, },
journal={IEICE TRANSACTIONS on Information},
title={End-to-End Delay Distribution on the Internet},
year={1999},
volume={E82-D},
number={4},
pages={762-768},
abstract={This paper proposes a new model which can approximate the delay time distribution in the Internet. It is well known that the delay time in communication links follows the exponential distribution. However, the earlier models cannot explain the distribution when a communication link is heavily overloaded. This paper proposes to use the M/M/S(m) model for the Internet. We have applied our model to the measurement results. This paper deals with one-way delay because it reflects the actual characteristics of communication links. Most measurement statistics in the Internet have been based on round-trip time delay between two end nodes. These characteristics are easily measured by sending sample packets from one node to the other. The receiver side echoes back the packets. However, the results are not always useful. A long distance communication link, such as a leased line, has two different fibers or wires for each direction: an incoming link, and an outgoing link. When the link is overloaded, the traffic in each link is quite different. The measurement of one-way delay is especially important for multimedia communications, because audio and video transmissions are essentially one-way traffic.},
keywords={},
doi={},
ISSN={},
month={April},}
Copiar
TY - JOUR
TI - End-to-End Delay Distribution on the Internet
T2 - IEICE TRANSACTIONS on Information
SP - 762
EP - 768
AU - Jun-ya KATO
AU - Atsuo SHIMIZU
AU - Shigeki GOTO
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E82-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 1999
AB - This paper proposes a new model which can approximate the delay time distribution in the Internet. It is well known that the delay time in communication links follows the exponential distribution. However, the earlier models cannot explain the distribution when a communication link is heavily overloaded. This paper proposes to use the M/M/S(m) model for the Internet. We have applied our model to the measurement results. This paper deals with one-way delay because it reflects the actual characteristics of communication links. Most measurement statistics in the Internet have been based on round-trip time delay between two end nodes. These characteristics are easily measured by sending sample packets from one node to the other. The receiver side echoes back the packets. However, the results are not always useful. A long distance communication link, such as a leased line, has two different fibers or wires for each direction: an incoming link, and an outgoing link. When the link is overloaded, the traffic in each link is quite different. The measurement of one-way delay is especially important for multimedia communications, because audio and video transmissions are essentially one-way traffic.
ER -