{"title":"Packet reordering, high speed networks and transport protocol performance","authors":"L. Gharai, C. Perkins, T. Lehman","doi":"10.1109/ICCCN.2004.1401591","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401591","url":null,"abstract":"We performed end-to-end measurements of UDP/IP flows across an Internet backbone network. Using this data, we characterized the packet reordering processes seen in the network. Our results demonstrate the high prevalence of packet reordering relative to packet loss, and show a strong correlation between packet rate and reordering on the network we studied. We conclude that, given the increased parallelism in modern networks and the demands of high performance applications, new application and protocol designs should treat packet reordering on an equal footing to packet loss, and must be robust and resilient to both in order to achieve high performance","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121913747","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":"A multicast protection algorithm for optical WDM networks","authors":"Y. Zhang, D. Sidhu","doi":"10.1109/ICCCN.2004.1401656","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401656","url":null,"abstract":"Supporting multicast applications has drawn increased attention in the development of WDM based optical networks. To provide a high level of availability in the face of node or link failure for multicast traffic, pre-assigned spare capacity in the networks is needed. This paper studies the problem of constructing a minimum-cost, two-connected subgraph, satisfying the wavelength conversion constraint, for a multicast request in a WDM network. It is known that the problem of finding the minimum cost of a two-connected graph is NP-hard. We prove that the problem of the optimal wavelength assignment for a given graph is also NP-hard. We propose a routing and wavelength assignment heuristic which aims not only to minimize the cost of the subgraph but also to reduce the number of wavelength conversions in the networks. Simulation experiments show that our proposed algorithm achieves performance close to the optimal solution when the average number of wavelengths in the networks is not low","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"97 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128002267","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":"Spatial multipath location aided ad hoc routing","authors":"Soumendra Nanda, R. Gray","doi":"10.1109/ICCCN.2004.1401731","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401731","url":null,"abstract":"Mobile ad-hoc networks are infrastructure-free networks of mobile nodes that communicate with each other wirelessly. There are several ad hoc routing algorithms at present that utilize position information (usually in two dimensional terms) to make routing decisions at each node. We consider routing algorithms that use 3D positional information, particularly a hybrid extension to LAR that works in 3D. We propose a new hierarchical, zone-based 3D routing algorithm, based on GRID by Liao, Tseng and Sheu (2001). Our new algorithm called \"hyper-GRID\" is a hybrid algorithm that uses multipath routing (alternate path caching) in 3D. We propose replacing LAR with multipath LAR (MLAR) in GRID. We have implemented MLAR and are validating MLAR through simulation using ns-2 and studying its efficiency, scalability and other properties. We use a random waypoint mobility model and compare our MLAR approach versus LAR, AODV and AOMDV in both 2D and 3D for a range of traffic and mobility scenarios","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134551552","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":"Analyzing the performance of Internet worm attack approaches","authors":"Wei Yu","doi":"10.1109/ICCCN.2004.1401717","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401717","url":null,"abstract":"Recent active worm propagation events show that active worms can spread in an automated fashion and flood the Internet in a very short period of time. In this paper, we address the issue by analyzing the impacts of worm attack approaches on Internet worm propagation. In particular, we analyze four worm attack approaches with considering following attack strategies: pure random-based/ PlP (peer-to-peer) hitlist-based strategies for victim selection and cooperation-based/ non-cooperation-based strategies for worm instance coordination. Our numerical analysis results show that the attack approach with combining the worm instance cooperation and P2P hitlist-based victim selection strategies achieves the best performance compared to all other attack approaches in term of attack efficiency","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131616024","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":"Optimal and guaranteed alternative LSP for multiple failures","authors":"Lemma Hundessa, J. Domingo-Pascual","doi":"10.1109/ICCCN.2004.1401587","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401587","url":null,"abstract":"Fast rerouting mechanisms are being studied in order to provide fault tolerance for LSP in an MPLS network in case of node or link failure. For QoS provision, protected LSP are handled apart from non-protected LSP. Rerouting mechanisms when applied on protected LSP must take into account QoS requirements. This paper presents a mechanism that is able to handle multiple failures along an LSP while using an optimal alternative LSP. The rerouting decision is taken close to the point of failure that reduces the restoration time. The use of pre-established alternative LSP also reduces the restoration time and avoids blocking when looking for an alternative path. The proposal is based on a hybrid approach of segment repair and path repair and provides a new alternative LSP using dynamic rerouting once the protected one has been rerouted. In this way, there is always an alternative LSP for each protected LSP. The over all performance (recovery time, end-to-end delay, packet losses, and network resource utilization) is compared with existing protection mechanisms by simulation. The proposed hybrid approach, optimal and guaranteed alternative path (OGAP), avoids the possible use of a non-optimal alternative LSP to reroute the protected traffic and provides the flexibility of alternative route selection and setup as well as better resource utilization. Moreover, our proposal guarantees at least one alternative LSF at any time for the traffic on the protected LSP","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129804932","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":"Model for series of reconfigurations in wavelength-routed optical networks","authors":"Passakon Prathombutr, J. Stach, E. Park, S. Tak","doi":"10.1109/ICCCN.2004.1401695","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401695","url":null,"abstract":"When a traffic demand is changed, a virtual topology could be reconfigured to serve that traffic and to retain high performance. We purpose a reconfiguration model that minimizes costly changes in a virtual topology and maximizes network performance for a series of reconfigurations in the long term. The model includes the reconfiguration process and the policy. The reconfiguration process finds a set of non-dominated solutions using the multi-objective evolutionary algorithm (MOEA) that optimizes two objectives by using the concept of Pareto optimal. The policy picks a solution from the set of solutions above using the Markov decision process (MDP). A case study based on simulation experiments is conducted to illustrate the application of our model","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130333627","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":"The impact of UMTS/WCDMA channel round-trip time variations on TCP Vegas throughput","authors":"A. Lo, G. Heijenk, I. Niemegeers","doi":"10.1109/ICCCN.2004.1401590","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401590","url":null,"abstract":"Universal Mobile Telecommunications System (UMTS) is a third-generation cellular network that enables high-speed mobile Internet access. This paper evaluates the performance of the transmission control protocol (TCP) Vegas and compares its performance with TCP Reno in a UMTS environment. Bulk data transfer was considered in the simulation with varying radio channel conditions. Data losses were assumed to be only caused by the radio channel. Simulation results show that the performance of Vegas is unexpectedly worse than Reno even though the data losses were completely recovered by the UMTS radio link control layer. This has led us to conduct a detailed investigation on the behavior of Vegas in order to identify the cause of performance degradation in Vegas. The poor performance of Vegas is attributed to the UMTS radio interface characteristics which characterize by large and highly variable TCP round-trip times. The round-trip time variations activated Vegas' congestion avoidance and the sender would interpret the round-trip time variation as a sign of network congestion. Consequently, the sender shrinks its window size which reduces the transmission rate even though no TCP packet losses occur. Furthermore, a sudden increase in the instantaneous round-trip time can trigger spurious timeouts at the TCP sender using Vegas which performs unnecessary retransmissions. Spurious timeouts can lead to significant throughput reduction. Reno, on the other hand, does not show any abnormality and delivers the expected performance","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132659545","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":"Admission control for variable bit rate traffic using variable service interval in IEEE 802.11e WLANs","authors":"Wing Fai Fan, D. Tsang, B. Bensaou","doi":"10.1109/ICCCN.2004.1401699","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401699","url":null,"abstract":"The IEEE 802.11 working group is currently working on the standard IEEE 802.11e and introduces the hybrid coordination function (HCF) to provide better QoS support to real-time traffic. A reference design of simple scheduling and admission control algorithm is proposed in a TGe consensus proposal. However, this scheduling and admission control unit only consider the mean data rate and mean packet size. The rate and packet size variation are not taken into account. Thus, it is only efficient to CBR traffic and the packet loss rate of VBR traffic may be very high. In W. F. Fan et al. (Aug. 2004), we analyzed the packet loss rate of the reference scheme and proposed a new method to determine the effective TXOP duration for admission control so that the packet loss rate of VBR flows can be guaranteed. In both reference and our proposed scheme, all stations use fixed schedule service interval (SI) which is the minimum of all maximum service interval of all admitted flows. Thus, maximum packet delay of all stations is limited by the most stringent SI and some traffic with larger delay bound may be over-guaranteed. Also, the efficiency of the admission control scheme in W. F. Fan et al. (Aug. 2004), becomes lower than that of the reference one. In this paper, we extend our admission control scheme by using variable service interval to improve the efficiency about 20% - 30% and avoid over guarantee on packet delay. Also, the packet loss rate of VBR traffic can be guaranteed","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129493583","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":"Path based routing algorithm for ad hoc networks","authors":"Weilin Zeng, T. Suda","doi":"10.1109/ICCCN.2004.1401581","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401581","url":null,"abstract":"In reactive routing protocols for ad hoc networks, the network wide flooding generates a significant amount of traffic overhead. The goal of This work is to reduce this overhead. We propose a path based route discovery scheme which can find a route to the destination with lower overhead than flooding-based route discovery. We also showed that the probability of finding a path to the destination by using path based routing request is not negligible. The path obtained through path based routing request may be far from optimal. Thus, we propose a route expanding scheme which can discover a better path based on the path discovered in path based routing discovery phase. By simulation, we showed that integrating these two schemes has better performance than existing reactive routing protocols like DSR.","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123929420","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":"A fast rerouting scheme for OSPF/IS-IS networks","authors":"Yong Liu, A. Reddy","doi":"10.1109/ICCCN.2004.1401585","DOIUrl":"https://doi.org/10.1109/ICCCN.2004.1401585","url":null,"abstract":"Most current backbone networks use link-state protocol, OSPF or IS-IS, as their intra-domain routing protocol. Link-state protocols perform global routing table update to route around the failures. It usually takes seconds. As real-time applications like VoIP emerge in recent years, there is a requirement for a fast rerouting mechanism to route around failures before all routers on the network update their routing tables. In addition, fast rerouting is more appropriate than global routing table update when failures are transient. We propose such a fast rerouting extension for link-state protocols. In our approach, when a link fails, the affected traffic is rerouted along a pre-computed rerouting path. In case rerouting cannot be done locally, the local router signals the minimal number of upstream routers to setup the rerouting path for rerouting. We propose algorithms that simplify the rerouting operation and the rerouting path setup. With a simple extension to the current link state protocols, our scheme can route around failures faster and involves minimal number of routers for rerouting.","PeriodicalId":229045,"journal":{"name":"Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117260237","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}