The growth of Internet eXchange Points (IXPs) as an important component in the Internet Autonomous System (AS) level topology has led to the need for a detailed evaluation of their effect on Internet topology evolution and inter-domain packet routing. In this work, we carry out initial measurements to analyze the effect of the additional peering links and observe path quality in terms of round-trip time and loss rates of routes going through an IXP between pairs of Internet hosts. We observe significant effects of IXP-links on Internet growth and topology evolution while the routing performance points to numerous cases of Triangle Inequality Violations (TIVs) due to the IXP switches.
{"title":"Understanding the impact of internet exchange points on internet topology and routing performance","authors":"M. Ahmad, R. Guha","doi":"10.1145/1921206.1921226","DOIUrl":"https://doi.org/10.1145/1921206.1921226","url":null,"abstract":"The growth of Internet eXchange Points (IXPs) as an important component in the Internet Autonomous System (AS) level topology has led to the need for a detailed evaluation of their effect on Internet topology evolution and inter-domain packet routing. In this work, we carry out initial measurements to analyze the effect of the additional peering links and observe path quality in terms of round-trip time and loss rates of routes going through an IXP between pairs of Internet hosts. We observe significant effects of IXP-links on Internet growth and topology evolution while the routing performance points to numerous cases of Triangle Inequality Violations (TIVs) due to the IXP switches.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"384 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126729809","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}
A. Abdelkefi, Yuming Jiang, Wen Wang, Arne Øslebø, O. Kvittem
Principal component analysis (PCA) is a statistical technique that has been used for data analysis and dimensionality reduction. It was introduced as a network traffic anomaly detection technique firstly in [1]. Since then, a lot of research attention has been received, which results in an extensive analysis and several extensions. In [2], the sensitivity of PCA to its tuning parameters, such as the dimension of the low-rank subspace and the detection threshold, on traffic anomaly detection was indicated. However, no explanation on the underlying reasons of the problem was given in [2]. In [3], further investigation on the PCA sensitivity was conducted and it was found that the PCA sensitivity comes from the inability of PCA to detect temporal correlations. Based on this finding, an extension of PCA to Kalman-Loeve expansion (KLE) was proposed in [3]. While KLE shows slight improvement, it still exhibits similar sensitivity issue since a new tuning parameter called temporal correlation range was introduced. Recently, in [4], additional effort was paid to illustrate the PCA-poisoning problem. To underline this problem, an evading strategy called Boiled-Frog was proposed which adds a high fraction of outliers to the traffic. To defend against this, the authors employed a more robust version of PCA called PCA-GRID. While PCA-GRID shows performance improvement regarding the robustness to the outliers, it experiences a high sensitivity to the threshold estimate and the k-dimensional subspace that maximizes the dispersion of the data. The purpose of this work is to consider another technique to address the PCA poisoning problems to provide robust traffic anomaly detection: The Principal Component Pursuit.
{"title":"Robust traffic anomaly detection with principal component pursuit","authors":"A. Abdelkefi, Yuming Jiang, Wen Wang, Arne Øslebø, O. Kvittem","doi":"10.1145/1921206.1921217","DOIUrl":"https://doi.org/10.1145/1921206.1921217","url":null,"abstract":"Principal component analysis (PCA) is a statistical technique that has been used for data analysis and dimensionality reduction. It was introduced as a network traffic anomaly detection technique firstly in [1]. Since then, a lot of research attention has been received, which results in an extensive analysis and several extensions. In [2], the sensitivity of PCA to its tuning parameters, such as the dimension of the low-rank subspace and the detection threshold, on traffic anomaly detection was indicated. However, no explanation on the underlying reasons of the problem was given in [2]. In [3], further investigation on the PCA sensitivity was conducted and it was found that the PCA sensitivity comes from the inability of PCA to detect temporal correlations. Based on this finding, an extension of PCA to Kalman-Loeve expansion (KLE) was proposed in [3]. While KLE shows slight improvement, it still exhibits similar sensitivity issue since a new tuning parameter called temporal correlation range was introduced. Recently, in [4], additional effort was paid to illustrate the PCA-poisoning problem. To underline this problem, an evading strategy called Boiled-Frog was proposed which adds a high fraction of outliers to the traffic. To defend against this, the authors employed a more robust version of PCA called PCA-GRID. While PCA-GRID shows performance improvement regarding the robustness to the outliers, it experiences a high sensitivity to the threshold estimate and the k-dimensional subspace that maximizes the dispersion of the data. The purpose of this work is to consider another technique to address the PCA poisoning problems to provide robust traffic anomaly detection: The Principal Component Pursuit.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124130275","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 broadcast nature of wireless networks is both a boon and a bane. On the one hand, multiple receivers may overhear a single transmission on the same channel at no additional cost above a point-to-point transmission; this property has been exploited in many opportunistic applications (e.g., [2]). On the other hand, the shared nature of the medium effectively forces wireless nodes to leverage frequency diversity and channelization for efficiency. Unfortunately, coordinating communication between nodes that may switch channels at any time is difficult and may in fact offset any performance gains from leveraging frequency diversity.
{"title":"Have your cake and eat it too!: enabling frequency diversity through opportunism","authors":"Edmund L. Wong, Sangmin Lee","doi":"10.1145/1921206.1921224","DOIUrl":"https://doi.org/10.1145/1921206.1921224","url":null,"abstract":"The broadcast nature of wireless networks is both a boon and a bane. On the one hand, multiple receivers may overhear a single transmission on the same channel at no additional cost above a point-to-point transmission; this property has been exploited in many opportunistic applications (e.g., [2]). On the other hand, the shared nature of the medium effectively forces wireless nodes to leverage frequency diversity and channelization for efficiency. Unfortunately, coordinating communication between nodes that may switch channels at any time is difficult and may in fact offset any performance gains from leveraging frequency diversity.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"232 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124252898","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}
Nadi Sarrar, A. Feldmann, S. Uhlig, Rob Sherwood, Xin Huang
We propose an alternative router design to fill the gap between PC-based open source software routers and commercial high-end routers. Our hardware accelerated software router uses commodity PC hardware running an open source software router for the control path, and couples it with programmable switching hardware by delegating most of the packet forwarding to the switch. We describe a prototype implementation and show that it is capable of handling the traffic requirements of an existing carrier aggregation network.
{"title":"Towards hardware accelerated software routers","authors":"Nadi Sarrar, A. Feldmann, S. Uhlig, Rob Sherwood, Xin Huang","doi":"10.1145/1921206.1921209","DOIUrl":"https://doi.org/10.1145/1921206.1921209","url":null,"abstract":"We propose an alternative router design to fill the gap between PC-based open source software routers and commercial high-end routers. Our hardware accelerated software router uses commodity PC hardware running an open source software router for the control path, and couples it with programmable switching hardware by delegating most of the packet forwarding to the switch. We describe a prototype implementation and show that it is capable of handling the traffic requirements of an existing carrier aggregation network.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"213 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121720155","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 address an issue relevant to Content-Centric Networks, in which content is discovered by flooding a request over the entire network. In our design, the collection of objects resolvable by a router is represented using Bloom filters. In response to the flooding, we send Bloom filters -- in essence transmitting all the object names represented within a router -- along with the object. In this manner, any subsequent request to an object represented within that Bloom filter will not need to be flooded. We present preliminary observations to validate our claims.
{"title":"Prefetching Bloom filters to control flooding in content-centric networks","authors":"A. W. Kazi","doi":"10.1145/1921206.1921230","DOIUrl":"https://doi.org/10.1145/1921206.1921230","url":null,"abstract":"We address an issue relevant to Content-Centric Networks, in which content is discovered by flooding a request over the entire network. In our design, the collection of objects resolvable by a router is represented using Bloom filters. In response to the flooding, we send Bloom filters -- in essence transmitting all the object names represented within a router -- along with the object. In this manner, any subsequent request to an object represented within that Bloom filter will not need to be flooded. We present preliminary observations to validate our claims.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121569129","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}
Depending on their configuration, administration, and provisioning, networks provide drastically different features. For instance, some networks provide little failure resilience while others provision failover capacity and deploy middleboxes to protect against denial of service attacks [1, 2]. Yet the standard IP interface masks these differences; every network appears to provide the same basic "dial-tone" service. Consequently, clients that desire certain network properties must resort to ad hoc techniques to detect these differences or must target the lowest common denominator service.
{"title":"NetQuery: a knowledge plane for reasoning about network properties","authors":"Alan Shieh, E. G. Sirer, F. Schneider","doi":"10.1145/1921206.1921231","DOIUrl":"https://doi.org/10.1145/1921206.1921231","url":null,"abstract":"Depending on their configuration, administration, and provisioning, networks provide drastically different features. For instance, some networks provide little failure resilience while others provision failover capacity and deploy middleboxes to protect against denial of service attacks [1, 2]. Yet the standard IP interface masks these differences; every network appears to provide the same basic \"dial-tone\" service. Consequently, clients that desire certain network properties must resort to ad hoc techniques to detect these differences or must target the lowest common denominator service.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114567070","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 present a new multipath transport protocol - Heterogeneous Multipath Transport Protocol (HMTP), powered by fountain codes to improve throughput performance and path utilization of heterogeneous multi-homing networks. HMTP solves the receive buffer blocking problem and eliminates the need for retransmission and in-order packet delivery, especially in a heterogeneous multi-homing network.
{"title":"Multipath transport protocol for heterogeneous multi-homing networks","authors":"YoungHa Hwang, B. O. Obele, Hyuk Lim","doi":"10.1145/1921206.1921212","DOIUrl":"https://doi.org/10.1145/1921206.1921212","url":null,"abstract":"We present a new multipath transport protocol - Heterogeneous Multipath Transport Protocol (HMTP), powered by fountain codes to improve throughput performance and path utilization of heterogeneous multi-homing networks. HMTP solves the receive buffer blocking problem and eliminates the need for retransmission and in-order packet delivery, especially in a heterogeneous multi-homing network.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125229914","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}
This paper describes our initial experiences with using the IP timestamp option to infer router statistics such as traffic shape and CPU load. By deducing this information through the use of an IP option, we can gather these statistics without administrative access to the router under study. This creates the potential for researchers to gather basic statistics from routers in wide-area networks which support this option.
{"title":"Inferring router statistics with IP timestamps","authors":"Andrew D. Ferguson, Rodrigo Fonseca","doi":"10.1145/1921206.1921218","DOIUrl":"https://doi.org/10.1145/1921206.1921218","url":null,"abstract":"This paper describes our initial experiences with using the IP timestamp option to infer router statistics such as traffic shape and CPU load. By deducing this information through the use of an IP option, we can gather these statistics without administrative access to the router under study. This creates the potential for researchers to gather basic statistics from routers in wide-area networks which support this option.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121639294","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 number of wireless devices and the traffic volume generated by these devices become significant today, and many devices begin supporting 802.11n protocol for higher-speed wireless access. However, the diversification in link types of end-hosts may degrade system performance. For example, hosts using 802.11 protocols had better not be relay nodes in a P2P live streaming system because 802.11 is a half-duplex protocol and usually less stable compared to modern wired links. Hence, understanding traffic characteristics of various link types is essential for improving or building network architectures. Moreover, estimation of the link type of a remote host possibly achieves better performance (e.g., higher throughput) in some network systems. Baiamonte et al. [1] have proposed an algorithm to detect wireless hosts from passive measurement by using the entropy of packet interarrival time (PIT). Wei et al. [2] have also proposed an algorithm to classify access network types. However, these algorithms pay no attention to new link types such as 802.11n, 3G, and WiMAX even though each of them has different characteristics and possibly degrades the performance of network systems.
如今，无线设备的数量和这些设备产生的通信量变得非常大，许多设备开始支持802.11n协议以实现更快的无线访问。但是，终端主机链路类型的多样化可能会降低系统性能。例如，使用802.11协议的主机最好不要成为P2P直播系统中的中继节点，因为802.11是一种半双工协议，与现代有线链路相比通常不太稳定。因此，了解各种链路类型的流量特征对于改进或构建网络架构至关重要。此外，在某些网络系统中，对远程主机的链路类型的估计可能实现更好的性能(例如，更高的吞吐量)。Baiamonte等人[1]提出了一种利用包间到达时间(packet interarrival time, PIT)熵从被动测量中检测无线主机的算法。Wei等人[2]也提出了一种接入网类型分类算法。然而，这些算法不考虑新的链路类型，如802.11n、3G和WiMAX，尽管它们具有不同的特征，可能会降低网络系统的性能。
{"title":"Towards characterization of wireless traffic in coexisting 802.11a/g and 802.11n network","authors":"H. Asai, K. Fukuda, H. Esaki","doi":"10.1145/1921206.1921208","DOIUrl":"https://doi.org/10.1145/1921206.1921208","url":null,"abstract":"The number of wireless devices and the traffic volume generated by these devices become significant today, and many devices begin supporting 802.11n protocol for higher-speed wireless access. However, the diversification in link types of end-hosts may degrade system performance. For example, hosts using 802.11 protocols had better not be relay nodes in a P2P live streaming system because 802.11 is a half-duplex protocol and usually less stable compared to modern wired links. Hence, understanding traffic characteristics of various link types is essential for improving or building network architectures. Moreover, estimation of the link type of a remote host possibly achieves better performance (e.g., higher throughput) in some network systems. Baiamonte et al. [1] have proposed an algorithm to detect wireless hosts from passive measurement by using the entropy of packet interarrival time (PIT). Wei et al. [2] have also proposed an algorithm to classify access network types. However, these algorithms pay no attention to new link types such as 802.11n, 3G, and WiMAX even though each of them has different characteristics and possibly degrades the performance of network systems.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114958386","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}
File Hosting Services (FHS) such as Rapidshare and Mega-upload have recently become popular. The decline of P2P file sharing has prompted various services including FHS to replace it. We propose a comprehensive multi-level characterization of the FHS ecosystem. We devise a measurement framework to collect datasets from multiple vantage points. To the best of our knowledge, this work is the first to characterize the FHS ecosystem. The work will highlight the content, usage, performance, infrastructure, and quality of service characteristics of FHS. FHS can have significant implications on Internet traffic, if these services were to supplant P2P as the dominant content sharing technology.
{"title":"Characterizing the file hosting service ecosystem","authors":"Aniket Mahanti, Niklas Carlsson, C. Williamson","doi":"10.1145/1921206.1921210","DOIUrl":"https://doi.org/10.1145/1921206.1921210","url":null,"abstract":"File Hosting Services (FHS) such as Rapidshare and Mega-upload have recently become popular. The decline of P2P file sharing has prompted various services including FHS to replace it. We propose a comprehensive multi-level characterization of the FHS ecosystem. We devise a measurement framework to collect datasets from multiple vantage points. To the best of our knowledge, this work is the first to characterize the FHS ecosystem. The work will highlight the content, usage, performance, infrastructure, and quality of service characteristics of FHS. FHS can have significant implications on Internet traffic, if these services were to supplant P2P as the dominant content sharing technology.","PeriodicalId":325024,"journal":{"name":"CoNEXT '10 Student Workshop","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127379770","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：