Mandjes et al 2007a - Revision history

Scipediacontent at 18:57, 5 March 2021

2021-03-05T18:57:13Z

Scipediacontent: Scipediacontent moved page Draft Content 917236267 to Mandjes et al 2007a

2021-02-11T13:26:53Z

Scipediacontent moved page Draft Content 917236267 to Mandjes et al 2007a

Scipediacontent: Created page with " == Abstract == xtabstractWeb measurements have shown that TCP flow sizes vary over several orders of magnitude. If network resources are shared fairly, the performance of sh..."

2021-02-11T13:26:51Z

Created page with " == Abstract == xtabstractWeb measurements have shown that TCP flow sizes vary over several orders of magnitude. If network resources are shared fairly, the performance of sh..."

New page

== Abstract ==

xtabstractWeb measurements have shown that TCP flow sizes vary over several orders of magnitude. If network resources are shared fairly, the performance of short TCP flows is seriously degraded by long flows. This motivates prioritization of short over long flows, leading to significant performance improvement for short flows, with very little degradation for long ones. By imposing an appropriate pricing structure, users can be provided with incentives so as to enforce such a prioritization in a distributed manner. We study the situation where users randomly generate elastic flows (according to a Poisson process), whose sizes have a general probability distribution. Users may choose between two service classes that share the available resources with preemptive priority for one class. We assume fair sharing within each class and use a processor-sharing model with two priority classes to evaluate flow transmission times. For given price functions, users choose the service that best fits their profile, or refrain from service (`balk') if the prices exceed the willingness to pay for both services. We show that under mild modeling assumptions, the revenue maximizing Nash equilibrium is such that short flows opt for the premium service and long flows for the low-priority service. It is optimal for medium-sized flows to balk when the offered load is relatively high. Although this allocation is the result of distributed control, aside from the flows that balk, it accomplishes precisely the prioritization of short flows which improves overall performance as mentioned above

== Original document ==

The different versions of the original document can be found in:

* [http://dx.doi.org/10.1016/j.orl.2006.03.018 http://dx.doi.org/10.1016/j.orl.2006.03.018]

* [https://dare.uva.nl/personal/pure/en/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(ae19bf2d-3ae6-473d-b435-d72a464a639b).html https://dare.uva.nl/personal/pure/en/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(ae19bf2d-3ae6-473d-b435-d72a464a639b).html]

* [https://ir.cwi.nl/pub/10839 https://ir.cwi.nl/pub/10839]

* [https://ir.cwi.nl/pub/10839/10839D.pdf https://ir.cwi.nl/pub/10839/10839D.pdf]

* [https://research.tue.nl/nl/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(2f537733-9dac-4b0f-b9ac-2cbace1138ef).html https://research.tue.nl/nl/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(2f537733-9dac-4b0f-b9ac-2cbace1138ef).html]

* [https://research.tue.nl/nl/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(de7beb4d-1323-4ed8-90b3-622db1b7a6c2).html https://research.tue.nl/nl/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(de7beb4d-1323-4ed8-90b3-622db1b7a6c2).html]

* [https://research.utwente.nl/en/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(b2a029f7-627c-4d5d-be09-73fc662ccdb8).html https://research.utwente.nl/en/publications/pricing-and-distributed-qos-control-for-elastic-network-traffic(b2a029f7-627c-4d5d-be09-73fc662ccdb8).html]

* [http://www.sciencedirect.com/science/article/pii/S0167637706000642 http://www.sciencedirect.com/science/article/pii/S0167637706000642],
: [http://dx.doi.org/10.1016/j.orl.2006.03.018 http://dx.doi.org/10.1016/j.orl.2006.03.018]

* [https://api.elsevier.com/content/article/PII:S0167637706000642?httpAccept=text/xml https://api.elsevier.com/content/article/PII:S0167637706000642?httpAccept=text/xml],
: [https://api.elsevier.com/content/article/PII:S0167637706000642?httpAccept=text/plain https://api.elsevier.com/content/article/PII:S0167637706000642?httpAccept=text/plain],
: [http://dx.doi.org/10.1016/j.orl.2006.03.018 http://dx.doi.org/10.1016/j.orl.2006.03.018] under the license https://www.elsevier.com/tdm/userlicense/1.0/

* [https://dblp.uni-trier.de/db/journals/orl/orl35.html#BergMN07 https://dblp.uni-trier.de/db/journals/orl/orl35.html#BergMN07],
: [https://www.narcis.nl/publication/RecordID/oai%3Apure.tue.nl%3Apublications%2F2f537733-9dac-4b0f-b9ac-2cbace1138ef https://www.narcis.nl/publication/RecordID/oai%3Apure.tue.nl%3Apublications%2F2f537733-9dac-4b0f-b9ac-2cbace1138ef],
: [https://www.sciencedirect.com/science/article/pii/S0167637706000642 https://www.sciencedirect.com/science/article/pii/S0167637706000642],
: [http://dx.doi.org/10.1016/j.orl.2006.03.018 http://dx.doi.org/10.1016/j.orl.2006.03.018],
: [https://ir.cwi.nl/pub/10839 https://ir.cwi.nl/pub/10839],
: [https://doi.org/10.1016/j.orl.2006.03.018 https://doi.org/10.1016/j.orl.2006.03.018],
: [https://dare.uva.nl/search?f1-organisation=Faculty%20of%20Science%20(FNWI)::Korteweg-de%20Vries%20Institute%20for%20Mathematics%20(KdVI);docsPerPage=1;startDoc=1 https://dare.uva.nl/search?f1-organisation=Faculty%20of%20Science%20(FNWI)::Korteweg-de%20Vries%20Institute%20for%20Mathematics%20(KdVI);docsPerPage=1;startDoc=1],
: [https://academic.microsoft.com/#/detail/2160423655 https://academic.microsoft.com/#/detail/2160423655]

@@ Line 2: / Line 2: @@
 == Abstract ==
-xtabstractWeb measurements have shown that TCP flow sizes vary over several orders of magnitude. If network resources are shared fairly, the performance of short TCP flows is seriously degraded by long flows. This motivates prioritization of short over long flows, leading to significant performance improvement for short flows, with very little degradation for long ones. By imposing an appropriate pricing structure, users can be provided with incentives so as to enforce such a prioritization in a distributed manner. We study the situation where users randomly generate elastic flows (according to a Poisson process), whose sizes have a general probability distribution. Users may choose between two service classes that share the available resources with preemptive priority for one class. We assume fair sharing within each class and use a processor-sharing model with two priority classes to evaluate flow transmission times. For given price functions, users choose the service that best fits their profile, or refrain from service (`balk') if the prices exceed the willingness to pay for both services. We show that under mild modeling assumptions, the revenue maximizing Nash equilibrium is such that short flows opt for the premium service and long flows for the low-priority service. It is optimal for medium-sized flows to balk when the offered load is relatively high. Although this allocation is the result of distributed control, aside from the flows that balk, it accomplishes precisely the prioritization of short flows which improves overall performance as mentioned above
+Web measurements have shown that TCP flow sizes vary over several orders of magnitude. If network resources are shared fairly, the performance of short TCP flows is seriously degraded by long flows. This motivates prioritization of short over long flows, leading to significant performance improvement for short flows, with very little degradation for long ones. By imposing an appropriate pricing structure, users can be provided with incentives so as to enforce such a prioritization in a distributed manner. We study the situation where users randomly generate elastic flows (according to a Poisson process), whose sizes have a general probability distribution. Users may choose between two service classes that share the available resources with preemptive priority for one class. We assume fair sharing within each class and use a processor-sharing model with two priority classes to evaluate flow transmission times. For given price functions, users choose the service that best fits their profile, or refrain from service (`balk') if the prices exceed the willingness to pay for both services. We show that under mild modeling assumptions, the revenue maximizing Nash equilibrium is such that short flows opt for the premium service and long flows for the low-priority service. It is optimal for medium-sized flows to balk when the offered load is relatively high. Although this allocation is the result of distributed control, aside from the flows that balk, it accomplishes precisely the prioritization of short flows which improves overall performance as mentioned above.
-−
 == Original document ==

← Older revision	Revision as of 13:26, 11 February 2021
(No difference)