Y. Timner, J. Pettersson, H. Hannu, Min Wang, I. Johansson
This work investigates rate adaptation of conversational video in a mobile system using Long Term Evolution, LTE, and where the adaptation is assisted by the radio network. The conventional way to do rate adaptation is through adaptation in the end-points where the transmission rate is selected based on measurements of received packets. This study investigates two network-based algorithms for rate adaptation, a rate fair algorithm that assigns the same rate to all conversational video users in a cell, and a resource fair algorithm that aims to give all users in the cell a fair amount of resources. Both algorithms are combined with delay based scheduling. Both network-based algorithms perform excellently. The delay stays low even when the resource utilization is close to 100%, and the video rate is adapted to the system load. As could be expected, the average user rates are higher with the resource fair algorithm. An end-point based adaptation algorithm is investigated as well, but it cannot keep a low delay at high load.
{"title":"Network assisted rate adaptation for conversational video over LTE, concept and performance evaluation","authors":"Y. Timner, J. Pettersson, H. Hannu, Min Wang, I. Johansson","doi":"10.1145/2630088.2630096","DOIUrl":"https://doi.org/10.1145/2630088.2630096","url":null,"abstract":"This work investigates rate adaptation of conversational video in a mobile system using Long Term Evolution, LTE, and where the adaptation is assisted by the radio network. The conventional way to do rate adaptation is through adaptation in the end-points where the transmission rate is selected based on measurements of received packets. This study investigates two network-based algorithms for rate adaptation, a rate fair algorithm that assigns the same rate to all conversational video users in a cell, and a resource fair algorithm that aims to give all users in the cell a fair amount of resources. Both algorithms are combined with delay based scheduling. Both network-based algorithms perform excellently. The delay stays low even when the resource utilization is close to 100%, and the video rate is adapted to the system load. As could be expected, the average user rates are higher with the resource fair algorithm. An end-point based adaptation algorithm is investigated as well, but it cannot keep a low delay at high load.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125378722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Network sharing is often hailed as a promising and cost-effective way to tackle the ever-increasing load of cellular networks. However, its actual effectiveness strongly depends on the correlation between the networks being joined -- intuitively, there is no benefit in joining two networks with exactly the same load and exactly the same deployment. In this paper, we analyse the deployment and traffic traces of two Irish operators to (i) study their correlation in space and time, and (ii) assess the potential benefit brought by network sharing. Through our analysis, we are able to show that network sharing is remarkably effective in making the load more regular over space, improving the operations and performance of cellular networks.
{"title":"Mobile network sharing between operators: a demand trace-driven study","authors":"Paolo Di Francesco, F. Malandrino, L. Dasilva","doi":"10.1145/2630088.2630090","DOIUrl":"https://doi.org/10.1145/2630088.2630090","url":null,"abstract":"Network sharing is often hailed as a promising and cost-effective way to tackle the ever-increasing load of cellular networks. However, its actual effectiveness strongly depends on the correlation between the networks being joined -- intuitively, there is no benefit in joining two networks with exactly the same load and exactly the same deployment. In this paper, we analyse the deployment and traffic traces of two Irish operators to (i) study their correlation in space and time, and (ii) assess the potential benefit brought by network sharing. Through our analysis, we are able to show that network sharing is remarkably effective in making the load more regular over space, improving the operations and performance of cellular networks.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122461213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Session details: Technical session 3: mobile networks","authors":"D. Kutscher","doi":"10.1145/3251060","DOIUrl":"https://doi.org/10.1145/3251060","url":null,"abstract":"","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126405012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We evaluate modern delay-based AQM algorithms in downstream DOCSIS 3.0 cable environments. Our focus is on fairness and application performance capabilities of two recently proposed delay-based AQM algorithms, CoDel and PIE. The evaluation involves scenarios that include tiered service levels and application workloads that include FTP, HTTP-based adaptive streaming, and VoIP traffic. Our results provide a snapshot of our current effort to evaluate AQM schemes that are likely to be deployed in emerging DOCSIS cable networks.
{"title":"Managing fairness and application performance with active queue management in docsis-based cable networks","authors":"James J. Martin, Gongbing Hong, J. Westall","doi":"10.1145/2630088.2630092","DOIUrl":"https://doi.org/10.1145/2630088.2630092","url":null,"abstract":"We evaluate modern delay-based AQM algorithms in downstream DOCSIS 3.0 cable environments. Our focus is on fairness and application performance capabilities of two recently proposed delay-based AQM algorithms, CoDel and PIE. The evaluation involves scenarios that include tiered service levels and application workloads that include FTP, HTTP-based adaptive streaming, and VoIP traffic. Our results provide a snapshot of our current effort to evaluate AQM schemes that are likely to be deployed in emerging DOCSIS cable networks.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134403903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
C. Paasch, Simone Ferlin Oliveira, Özgü Alay, O. Bonaventure
Today many end hosts are equipped with multiple interfaces. These interfaces can be utilized simultaneously by multipath protocols to pool resources of the links in an efficient way while also providing resilience to eventual link failures. However how to schedule the data segments over multiple links is a challenging problem, and highly influences the performance of multipath protocols. In this paper, we focus on different schedulers for Multipath TCP. We first design and implement a generic modular scheduler framework that enables testing of different schedulers for Multipath TCP. We then use this framework to do an in-depth analysis of different schedulers by running emulated and real-world experiments on a testbed. We consider bulk data transfer as well as application limited traffic and identify metrics to quantify the scheduler's performance. Our results shed light on how scheduling decisions can help to improve multipath transfer.
{"title":"Experimental evaluation of multipath TCP schedulers","authors":"C. Paasch, Simone Ferlin Oliveira, Özgü Alay, O. Bonaventure","doi":"10.1145/2630088.2631977","DOIUrl":"https://doi.org/10.1145/2630088.2631977","url":null,"abstract":"Today many end hosts are equipped with multiple interfaces. These interfaces can be utilized simultaneously by multipath protocols to pool resources of the links in an efficient way while also providing resilience to eventual link failures. However how to schedule the data segments over multiple links is a challenging problem, and highly influences the performance of multipath protocols. In this paper, we focus on different schedulers for Multipath TCP. We first design and implement a generic modular scheduler framework that enables testing of different schedulers for Multipath TCP. We then use this framework to do an in-depth analysis of different schedulers by running emulated and real-world experiments on a testbed. We consider bulk data transfer as well as application limited traffic and identify metrics to quantify the scheduler's performance. Our results shed light on how scheduling decisions can help to improve multipath transfer.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120999253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Welcome to the Capacity Sharing Workshop (CSWS) at ACM SIGCOMM 2014. Today's Internet traffic consists of a wide variety of application and services that have very different requirements on the Quality of Service (QoS) the network should provide. The continuously growing demand for bandwidth (massive growth of mobile network user population and increasing data rates) and the fast increasing deployment of virtualization (in data centers, mobile core network, access networks) make effective and efficient capacity sharing a key element for managing cost as well of quality of experience for users. The objective of CSWS is to provide a forum for exchanging recent research results and new ideas addressing these issues -- focusing on two research areas: (Per-user) fair and (economically) efficient capacity sharing including generalization of data center techniques to Internet; and Support of application requirements, especially low latency, and measurements/measurement techniques for assessing the current Qos/QoE in the Internet. CSWS-2014 provides a forum for sharing research and experimentation results from a number of current (sub-)activities in the area of capacity sharing. This includes the bufferbloat initiative (aiming at reducing excessive delay caused by over-dimensioned and badly managed buffers in network equipment), new work on active queue management (such as CoDel and PIE), new approaches to better expose, respond to and police congestion (Congestion Exposure, MPTCP, Data Center TCP), and congestion control/mitigation for real-time media streams (IETF RMCAT). With the help of an excellent technical program committee of 39 international researchers, we finally selected 9 interesting papers for the workshop program that, as we believe, not only address important current research topics in the fields but are also suitable to create interesting and fruitful discussions at the workshop. Moreover, the workshop features a keynote by Dave Taht, a networking researcher and developer, known for his contributions to overcome Bufferbloat. The technical program is structured into three topic blocks on 1) design and evaluation of Queuing and Scheduling mechanisms, 2) new extensions to Transport Protocols, and 3) rate adaptation and bandwidth allocation strategies in Mobile Networks.
{"title":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","authors":"M. Kühlewind, D. Kutscher","doi":"10.1145/2630088","DOIUrl":"https://doi.org/10.1145/2630088","url":null,"abstract":"Welcome to the Capacity Sharing Workshop (CSWS) at ACM SIGCOMM 2014. \u0000 \u0000Today's Internet traffic consists of a wide variety of application and services that have very different requirements on the Quality of Service (QoS) the network should provide. The continuously growing demand for bandwidth (massive growth of mobile network user population and increasing data rates) and the fast increasing deployment of virtualization (in data centers, mobile core network, access networks) make effective and efficient capacity sharing a key element for managing cost as well of quality of experience for users. \u0000 \u0000The objective of CSWS is to provide a forum for exchanging recent research results and new ideas addressing these issues -- focusing on two research areas: \u0000(Per-user) fair and (economically) efficient capacity sharing including generalization of data center techniques to Internet; and \u0000Support of application requirements, especially low latency, and measurements/measurement techniques for assessing the current Qos/QoE in the Internet. \u0000 \u0000 \u0000 \u0000CSWS-2014 provides a forum for sharing research and experimentation results from a number of current (sub-)activities in the area of capacity sharing. This includes the bufferbloat initiative (aiming at reducing excessive delay caused by over-dimensioned and badly managed buffers in network equipment), new work on active queue management (such as CoDel and PIE), new approaches to better expose, respond to and police congestion (Congestion Exposure, MPTCP, Data Center TCP), and congestion control/mitigation for real-time media streams (IETF RMCAT). \u0000 \u0000With the help of an excellent technical program committee of 39 international researchers, we finally selected 9 interesting papers for the workshop program that, as we believe, not only address important current research topics in the fields but are also suitable to create interesting and fruitful discussions at the workshop. \u0000 \u0000Moreover, the workshop features a keynote by Dave Taht, a networking researcher and developer, known for his contributions to overcome Bufferbloat. The technical program is structured into three topic blocks on 1) design and evaluation of Queuing and Scheduling mechanisms, 2) new extensions to Transport Protocols, and 3) rate adaptation and bandwidth allocation strategies in Mobile Networks.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117014951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Bufferbloat project was founded three and half years ago to explore solutions to why modern data networks became so slow when loaded. We set out to explore the literature, and code, from the earliest days of the Internet to today, to try and find out what had gone wrong, and to find ways to fix it. A result of that effort (so far!) has been a renaissance in interest in congestion control, and huge - orders of magnitude - improvements in network latency along the edge of the Internet, with deployable new algorithms and code. Surprisingly, we also made large improvements in simultaneous bidirectional goodput on asymmetric networks. This talk goes into the history and future of Active Queue Management (AQM) and fair queuing algorithms, touches upon current work across the field and tries to identify useful techniques for exploring and designing future work that can scale.
{"title":"The value of repeatable experiments and negative results: - a journey through the history and future of AQM and fair queuing algorithms.","authors":"Dave Täht","doi":"10.1145/2630088.2652480","DOIUrl":"https://doi.org/10.1145/2630088.2652480","url":null,"abstract":"The Bufferbloat project was founded three and half years ago to explore solutions to why modern data networks became so slow when loaded. We set out to explore the literature, and code, from the earliest days of the Internet to today, to try and find out what had gone wrong, and to find ways to fix it. A result of that effort (so far!) has been a renaissance in interest in congestion control, and huge - orders of magnitude - improvements in network latency along the edge of the Internet, with deployable new algorithms and code. Surprisingly, we also made large improvements in simultaneous bidirectional goodput on asymmetric networks. This talk goes into the history and future of Active Queue Management (AQM) and fair queuing algorithms, touches upon current work across the field and tries to identify useful techniques for exploring and designing future work that can scale.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125974266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We use ns-2 simulations to compare RED's gentle_ mode to CoDel in terms of their ability to reduce the latency for various TCP variants. We use a common dumbbell topol- ogy with Pareto background traffic, and measure the packet delays and transmission time of a 10 MB FTP transfer. In our scenarios, we find that CoDel reduces the latency by 87%, but RED still manages to reduce it by 75%. However, the use of CoDel results in a transmission time 42% longer than when using RED. In light of its maturity, we therefore argue that RED could be considered as a good candidate to tackle Bufferbloat.
{"title":"Revisiting old friends: is CoDel really achieving what RED cannot?","authors":"N. Kuhn, E. Lochin, O. Mehani","doi":"10.1145/2630088.2630094","DOIUrl":"https://doi.org/10.1145/2630088.2630094","url":null,"abstract":"We use ns-2 simulations to compare RED's gentle_ mode to CoDel in terms of their ability to reduce the latency for various TCP variants. We use a common dumbbell topol- ogy with Pareto background traffic, and measure the packet delays and transmission time of a 10 MB FTP transfer. In our scenarios, we find that CoDel reduces the latency by 87%, but RED still manages to reduce it by 75%. However, the use of CoDel results in a transmission time 42% longer than when using RED. In light of its maturity, we therefore argue that RED could be considered as a good candidate to tackle Bufferbloat.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133658575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The paper proposes ConEx Lite to ease the deployment of congestion bitrate policing in existing 4G and WiFi mobile networks with initial objectives to improve end user experience and backhaul network performance and dimensioning. ConEx Lite is a congestion management solution working at the bearer or tunnel layer independent from UE terminals and Internet endpoints or other transport protocol (e.g. TCP) implementations. It consists of simple functions that can be implemented on existing radio access and core nodes without negatively impacting the performance of the mobile network. ConEx Lite provides faster response to congestion and allows mobile operators to control the congestion volume policies according to their radio access technology and/or service mix.
{"title":"Conex lite for mobile networks","authors":"Steve Baillargeon, I. Johansson","doi":"10.1145/2630088.2630091","DOIUrl":"https://doi.org/10.1145/2630088.2630091","url":null,"abstract":"The paper proposes ConEx Lite to ease the deployment of congestion bitrate policing in existing 4G and WiFi mobile networks with initial objectives to improve end user experience and backhaul network performance and dimensioning. ConEx Lite is a congestion management solution working at the bearer or tunnel layer independent from UE terminals and Internet endpoints or other transport protocol (e.g. TCP) implementations. It consists of simple functions that can be implemented on existing radio access and core nodes without negatively impacting the performance of the mobile network. ConEx Lite provides faster response to congestion and allows mobile operators to control the congestion volume policies according to their radio access technology and/or service mix.","PeriodicalId":106412,"journal":{"name":"Proceedings of the 2014 ACM SIGCOMM workshop on Capacity sharing workshop","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130402929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：