Pub Date : 2020-12-01DOI: 10.1109/ICNP.2010.5762780
T. Higashino, Kenji Suzuki
{"title":"Message from the General Co-Chairs","authors":"T. Higashino, Kenji Suzuki","doi":"10.1109/ICNP.2010.5762780","DOIUrl":"https://doi.org/10.1109/ICNP.2010.5762780","url":null,"abstract":"","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"58 11","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91440832","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}
Pub Date : 2017-11-21DOI: 10.1109/ICNP.2017.8117592
Mu He, Patrick Kalmbach, Andreas Blenk, W. Kellerer, S. Schmid
This paper is motivated by the emerging vision of an automated and data-driven optimization of communication networks, making it possible to fully exploit the flexibilities offered by modern network technologies and heralding an era of fast and self-adjusting networks. We build upon our recent study of machine-learning approaches to (statically) optimize resource allocations based on the data produced by network algorithms in the past. We take our study a crucial step further by considering dynamic scenarios: scenarios where communication patterns can change over time. In particular, we investigate network algorithms which learn from the traffic distribution (the feature vector), in order to predict global network allocations (a multi-label problem). As a case study, we consider a well-studied fc-median problem arising in Software-Defined Networks, and aim to imitate and speedup existing heuristics as well as to predict good initial solutions for local search algorithms. We compare different machine learning algorithms by simulation and find that neural network can provide the best abstraction, saving up to two-thirds of the algorithm runtime.
{"title":"Algorithm-data driven optimization of adaptive communication networks","authors":"Mu He, Patrick Kalmbach, Andreas Blenk, W. Kellerer, S. Schmid","doi":"10.1109/ICNP.2017.8117592","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117592","url":null,"abstract":"This paper is motivated by the emerging vision of an automated and data-driven optimization of communication networks, making it possible to fully exploit the flexibilities offered by modern network technologies and heralding an era of fast and self-adjusting networks. We build upon our recent study of machine-learning approaches to (statically) optimize resource allocations based on the data produced by network algorithms in the past. We take our study a crucial step further by considering dynamic scenarios: scenarios where communication patterns can change over time. In particular, we investigate network algorithms which learn from the traffic distribution (the feature vector), in order to predict global network allocations (a multi-label problem). As a case study, we consider a well-studied fc-median problem arising in Software-Defined Networks, and aim to imitate and speedup existing heuristics as well as to predict good initial solutions for local search algorithms. We compare different machine learning algorithms by simulation and find that neural network can provide the best abstraction, saving up to two-thirds of the algorithm runtime.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"37 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91085754","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}
Pub Date : 2017-10-13DOI: 10.1109/ICNP.2017.8117579
P. Chuprikov, A. Davydow, Kirill Kogan, S. Nikolenko, Alexander Sirotkin
Efficient representation of data aggregations is a fundamental problem in modern big data applications. We present a formalization of compute-aggregate planning parameterized by the aggregation function.
{"title":"Planning in compute-aggregate problems as optimization problems on graphs","authors":"P. Chuprikov, A. Davydow, Kirill Kogan, S. Nikolenko, Alexander Sirotkin","doi":"10.1109/ICNP.2017.8117579","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117579","url":null,"abstract":"Efficient representation of data aggregations is a fundamental problem in modern big data applications. We present a formalization of compute-aggregate planning parameterized by the aggregation function.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"17 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2017-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75207189","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}
Pub Date : 2017-10-10DOI: 10.1109/ICNP.2017.8117533
Kirill Kogan, Danushka Menikkumbura, G. Petri, Y. Noh, S. Nikolenko, Alexander Sirotkin, P. Eugster
Buffering architectures and policies for their efficient management constitute one of the core ingredients of a network architecture. However, despite strong incentives to experiment with, and deploy, new policies, the opportunities for alterating anything beyond minor elements of such policies are limited. In this work we introduce a new specification language, OpenQueue, that allows users to specify entire buffering architectures and policies conveniently through several comparators and simple functions. We show examples of buffer management policies in OpenQueue and empirically demonstrate its direct impact on performance in various settings.
{"title":"A programmable buffer management platform","authors":"Kirill Kogan, Danushka Menikkumbura, G. Petri, Y. Noh, S. Nikolenko, Alexander Sirotkin, P. Eugster","doi":"10.1109/ICNP.2017.8117533","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117533","url":null,"abstract":"Buffering architectures and policies for their efficient management constitute one of the core ingredients of a network architecture. However, despite strong incentives to experiment with, and deploy, new policies, the opportunities for alterating anything beyond minor elements of such policies are limited. In this work we introduce a new specification language, OpenQueue, that allows users to specify entire buffering architectures and policies conveniently through several comparators and simple functions. We show examples of buffer management policies in OpenQueue and empirically demonstrate its direct impact on performance in various settings.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"20 1","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2017-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87090197","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}
Pub Date : 2017-10-10DOI: 10.1109/ICNP.2017.8117542
P. Chuprikov, Kirill Kogan, S. Nikolenko
Ternary Content-Addressable Memory (tcam) is a powerful tool to represent network services with line-rate lookup time. There are various software-based approaches to represent multi-field packet classifiers. Unfortunately, all of them either require exponential memory or apply additional constraints on field representations (e.g, prefixes or exact values) to have line-rate lookup time. In this work, we propose alternatives to tcam and introduce a novel approach to represent packet classifiers based on ternary bit strings (without constraining field representation) on commodity longest-prefix-match (lpm) infrastructures. These representations are built on a novel property, prefix reorderability, that defines how to transform an ordered set of ternary bit strings to prefixes with lpm priorities in linear memory. Our results are supported by evaluations on large-scale packet classifiers with real parameters from ClassBench; moreover, we have developed a prototype in P4 to support these types of transformations.
{"title":"General ternary bit strings on commodity longest-prefix-match infrastructures","authors":"P. Chuprikov, Kirill Kogan, S. Nikolenko","doi":"10.1109/ICNP.2017.8117542","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117542","url":null,"abstract":"Ternary Content-Addressable Memory (tcam) is a powerful tool to represent network services with line-rate lookup time. There are various software-based approaches to represent multi-field packet classifiers. Unfortunately, all of them either require exponential memory or apply additional constraints on field representations (e.g, prefixes or exact values) to have line-rate lookup time. In this work, we propose alternatives to tcam and introduce a novel approach to represent packet classifiers based on ternary bit strings (without constraining field representation) on commodity longest-prefix-match (lpm) infrastructures. These representations are built on a novel property, prefix reorderability, that defines how to transform an ordered set of ternary bit strings to prefixes with lpm priorities in linear memory. Our results are supported by evaluations on large-scale packet classifiers with real parameters from ClassBench; moreover, we have developed a prototype in P4 to support these types of transformations.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"940 ","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2017-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72551515","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}
Pub Date : 2017-10-01DOI: 10.1109/ICNP.2017.8117569
Shuo Wang, Jiao Zhang, Tao Huang, Tian Pan, Jiang Liu, Yun-jie Liu
ECN thresholds have limited operational range and very strict scope. Lower thresholds exacerbate the queue underflow while higher thresholds increase the queueing delays. In this paper, an Adaptive ECN (A-ECN) marking scheme is proposed to enhance the performance of ECN. A-ECN can adaptively adjust ECN marking thresholds in different scenarios to achieve good generality. Therefore, network operators can directly deploy A-ECN in various environments regardless of underlying queue types and bandwidth.
{"title":"Adaptively adjusting ECN marking thresholds for datacenter networks","authors":"Shuo Wang, Jiao Zhang, Tao Huang, Tian Pan, Jiang Liu, Yun-jie Liu","doi":"10.1109/ICNP.2017.8117569","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117569","url":null,"abstract":"ECN thresholds have limited operational range and very strict scope. Lower thresholds exacerbate the queue underflow while higher thresholds increase the queueing delays. In this paper, an Adaptive ECN (A-ECN) marking scheme is proposed to enhance the performance of ECN. A-ECN can adaptively adjust ECN marking thresholds in different scenarios to achieve good generality. Therefore, network operators can directly deploy A-ECN in various environments regardless of underlying queue types and bandwidth.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"176 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85955571","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}
Pub Date : 2017-10-01DOI: 10.1109/ICNP.2017.8117554
M. T. Raza, Dongho Kim, Kyu-Han Kim, Songwu Lu, M. Gerla
LTE Network Function Virtualization (LTE-NFV) scales user services in a low cost fashion by transforming the centralized legacy LTE Core architecture to a distributed architecture. This distributed architecture makes multiple instances of LTE Network Functions (NFs) and virtualizes them on commodity data-center network. The functionality of LTE-NFV architecture breaks however, since the distributed NF instances connected via unreliable IP links delay the execution of critical events. The failure of time-critical events results in users' quality of service degradation and temporary service unavailability. In this paper, we propose a new way to virtualize LTE core network. We argue that logic-based NFs segregation should be done for NFV, instead of instance-based NFs segregation done in current NFV implementation. Our approach of ‘logic-based NFs segregation’ combines the logic of an event into a single NF, thus localizing the execution of critical events to one virtual machine. This way, only the localized entities exchange signalling messages, and the events do not experience large delays. We further reduce the delays by exploiting the parallelism in LTE network protocols; and partition these protocols such that their signalling messages run in parallel. In addition, we eliminate unnecessary messages to reduce the signalling overhead. We build our system prototype over OpenEPC LTE core network in virtualized platform. Our results show that we can reduce event execution time and signalling overhead up to 50% and 40%, respectively.
{"title":"Rethinking LTE network functions virtualization","authors":"M. T. Raza, Dongho Kim, Kyu-Han Kim, Songwu Lu, M. Gerla","doi":"10.1109/ICNP.2017.8117554","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117554","url":null,"abstract":"LTE Network Function Virtualization (LTE-NFV) scales user services in a low cost fashion by transforming the centralized legacy LTE Core architecture to a distributed architecture. This distributed architecture makes multiple instances of LTE Network Functions (NFs) and virtualizes them on commodity data-center network. The functionality of LTE-NFV architecture breaks however, since the distributed NF instances connected via unreliable IP links delay the execution of critical events. The failure of time-critical events results in users' quality of service degradation and temporary service unavailability. In this paper, we propose a new way to virtualize LTE core network. We argue that logic-based NFs segregation should be done for NFV, instead of instance-based NFs segregation done in current NFV implementation. Our approach of ‘logic-based NFs segregation’ combines the logic of an event into a single NF, thus localizing the execution of critical events to one virtual machine. This way, only the localized entities exchange signalling messages, and the events do not experience large delays. We further reduce the delays by exploiting the parallelism in LTE network protocols; and partition these protocols such that their signalling messages run in parallel. In addition, we eliminate unnecessary messages to reduce the signalling overhead. We build our system prototype over OpenEPC LTE core network in virtualized platform. Our results show that we can reduce event execution time and signalling overhead up to 50% and 40%, respectively.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"101 1","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77041062","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}
In some special circumstances, e.g. tsunamis, floods, battlefields, earthquakes, etc., communication infrastructures are damaged or non-existent, as well as unmanned aerial vehicle (UAV) cluster. For the communication between people or UAVs, UAVs or mobile smart devices (MSDs) can be used to construct Mobile Ad Hoc Networks (MANETs), and Multipath TCP (MPTCP) can be used to simultaneously transmit in one TCP connection via multiple interfaces of MSDs. However the original MPTCP subpaths creating algorithm can establish multiple subpaths between two adjacent nodes, thus cannot achieve true concurrent data transmission. To solve this issue, we research and improve both the algorithm of adding routing table entries and the algorithm of establishing subpaths to offer more efficient use of multiple subpaths and better network traffic load balancing. The main works are as follows: (1) improve multi-hop routing protocol; (2) run MPTCP on UAVs or MSDs; (3) improve MPTCP subpaths establishment algorithm. The results show that our algorithms have better performance than the original MPTCP in achieving higher data throughput.
{"title":"The implementation of improved MPTCP in MANETs","authors":"Tongguang Zhang, Shuai Zhao, Yulong Shi, Bingfei Ren, B. Cheng, Junliang Chen","doi":"10.1109/ICNP.2017.8117583","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117583","url":null,"abstract":"In some special circumstances, e.g. tsunamis, floods, battlefields, earthquakes, etc., communication infrastructures are damaged or non-existent, as well as unmanned aerial vehicle (UAV) cluster. For the communication between people or UAVs, UAVs or mobile smart devices (MSDs) can be used to construct Mobile Ad Hoc Networks (MANETs), and Multipath TCP (MPTCP) can be used to simultaneously transmit in one TCP connection via multiple interfaces of MSDs. However the original MPTCP subpaths creating algorithm can establish multiple subpaths between two adjacent nodes, thus cannot achieve true concurrent data transmission. To solve this issue, we research and improve both the algorithm of adding routing table entries and the algorithm of establishing subpaths to offer more efficient use of multiple subpaths and better network traffic load balancing. The main works are as follows: (1) improve multi-hop routing protocol; (2) run MPTCP on UAVs or MSDs; (3) improve MPTCP subpaths establishment algorithm. The results show that our algorithms have better performance than the original MPTCP in achieving higher data throughput.","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"69 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84165904","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}
Pub Date : 2017-10-01DOI: 10.1109/ICNP.2017.8117548
Cheng Zhang, J. Bi, Yu Zhou, Jianping Wu, Bingyang Liu, Zhaogeng Li, A. B. Dogar, Yangyang Wang
While extending network programmability to a larger degree, P4 also raises the risks of incurring runtime bugs after the deployment of P4 programs. These runtime bugs, if not handled promptly and properly, can ruin the functionality and performance of networks. Unfortunately, the absence of runtime debuggers makes troubleshooting of P4 program bugs challenging and intricate for operators. This paper is devoted to the on-the-fly debugging of runtime bugs in P4-enabled networks. We propose P4DB, a general debugging platform that empowers operators to debug P4 programs in three levels of visibility by provisioning operator-friendly primitives. By P4DB, operators can use the watch primitive to quickly narrow the debugging scope from network level or device level to table level, then use the break and next primitives to decompose the match-action table into three steps and troubleshoot the runtime bugs step by step. We implemented a prototype of P4DB and evaluated the performance in terms of the data plane, control plane and control channel. On P4-specific programmable data plane, P4DB merely introduces a small throughput penalty (1.3%∼13.8%) and imposes a little-increased delay (0.6%∼11.9%).
{"title":"P4DB: On-the-fly debugging of the programmable data plane","authors":"Cheng Zhang, J. Bi, Yu Zhou, Jianping Wu, Bingyang Liu, Zhaogeng Li, A. B. Dogar, Yangyang Wang","doi":"10.1109/ICNP.2017.8117548","DOIUrl":"https://doi.org/10.1109/ICNP.2017.8117548","url":null,"abstract":"While extending network programmability to a larger degree, P4 also raises the risks of incurring runtime bugs after the deployment of P4 programs. These runtime bugs, if not handled promptly and properly, can ruin the functionality and performance of networks. Unfortunately, the absence of runtime debuggers makes troubleshooting of P4 program bugs challenging and intricate for operators. This paper is devoted to the on-the-fly debugging of runtime bugs in P4-enabled networks. We propose P4DB, a general debugging platform that empowers operators to debug P4 programs in three levels of visibility by provisioning operator-friendly primitives. By P4DB, operators can use the watch primitive to quickly narrow the debugging scope from network level or device level to table level, then use the break and next primitives to decompose the match-action table into three steps and troubleshoot the runtime bugs step by step. We implemented a prototype of P4DB and evaluated the performance in terms of the data plane, control plane and control channel. On P4-specific programmable data plane, P4DB merely introduces a small throughput penalty (1.3%∼13.8%) and imposes a little-increased delay (0.6%∼11.9%).","PeriodicalId":6462,"journal":{"name":"2017 IEEE 25th International Conference on Network Protocols (ICNP)","volume":"34 1","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86470000","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：