dc.rights.license | http://creativecommons.org/licenses/by-nc-sa/3.0/ve/ | |
dc.contributor.author | Arcia Moret, Andrés Emilio | |
dc.contributor.author | Montavont, Nicolas | |
dc.contributor.author | Bonnin, Jean Marie | |
dc.contributor.author | Ros, David | |
dc.date.accessioned | 2010-05-12T15:38:27Z | |
dc.date.available | 2010-05-12T15:38:27Z | |
dc.date.issued | 2010-05-12T15:38:27Z | |
dc.identifier.uri | http://www.saber.ula.ve/handle/123456789/30997 | |
dc.description.abstract | In TCP, receivers usually delay the emission of acknowledgements (ACK) packets for efficiency purposes (e.g., alleviate the processors charge or piggyback information in telnet connections). However, just as a TCP receiver may send less than one ACK per incoming data packet, it might also send more than one ACK per data packet without breaking the fundamental ACK semantics. In this article we investigate the impact of systematically increasing the ACK frequency and we discuss the uses and misuses of the technique. Interestingly, even when the ACK division seems applicable to unfairly gain bandwidth, results are not straightforward. There are several considerations that limit the impact of ACK division, such as the interactions with link layer protocols, the inners of TCP, the background traffic and the TCP congestion control algorithms. | es_VE |
dc.language.iso | es | es_VE |
dc.rights | info:eu-repo/semantics/openAccess | |
dc.subject | ACK division | es_VE |
dc.subject | Congestion control | es_VE |
dc.subject | TCP | es_VE |
dc.subject | CWND | es_VE |
dc.title | TCP ACK division revisited | es_VE |
dc.type | info:eu-repo/semantics/article | |
dc.description.colacion | 1-5 | es_VE |
dc.description.email | amoret@ula.ve | es_VE |
dc.description.email | david.ros@telecom-bretagne.eu | es_VE |
dc.subject.facultad | Facultad de Ingeniería | es_VE |
dc.subject.thematiccategory | Ingeniería | es_VE |
dc.subject.tipo | Artículos | es_VE |
dc.type.media | Texto | es_VE |