Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not ...Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.展开更多
Packet classification on multi-fields is a fundamental mechanism in network equipments,and various classification solutions have been proposed.Because of inherent difficulties,many of these solutions scale poorly in e...Packet classification on multi-fields is a fundamental mechanism in network equipments,and various classification solutions have been proposed.Because of inherent difficulties,many of these solutions scale poorly in either time or space as rule sets grow in size.Recursive Flow Classification（RFC） is an algorithm with a very high classifying speed. However,its preprocessing complexity and memory requirement are rather high.In this paper,we propose an enhanced RFC（ERFC） algorithm,in which a hash-based aggregated bit vector scheme is exploited to speed up its preprocessing procedure.A compressed and cacheable data structure is also introduced to decrease total memory requirement and improve its searching performance.Evaluation results show that ERFC provides a great improvement over RFC in both space requirement and preprocessing time.The search time complexity of ERFC is equivalent to that of RFC in the worst case; and its average classifying speed is improved by about 100%.展开更多
According to the X/Y flow classification method based on TCP and UDP port, a new method named self adaptive X/Y flow classification method is proposed in the paper, which can make the curve of the ratio of la...According to the X/Y flow classification method based on TCP and UDP port, a new method named self adaptive X/Y flow classification method is proposed in the paper, which can make the curve of the ratio of label resource usage more stable than ever so as to improve the performance of both L3 forwarding and L2 label switching of LER in MPLS networks. With the simulation of real Internet data, a satisfactory classification result has been obtained.展开更多
Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous deb...Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous debris flows showing complicated sediment composition and sedimentary processes were poorly understood. The main objective of this work is to establish a classification scheme and facies sequence models of subaqueous debris flows for well understanding their sedimentary processes and depositional characteristics.展开更多
基金supported by the National Natural Science Foundation of China(61962016)the Ministry of Science and Technology of China(G2022033002L)+1 种基金National Natural Science Foundation of Guangxi(2022JJA170057)Guangxi Education Department’s Project on Improving the Basic Research Ability of Young and Middleaged Teachers in Universities(2023ky0812,Research on Statistical Network Delay Predictions in Large-scale SDNs).
文摘Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.
基金Supported by the National Basic Research 973 Program of China under Grant No.2009CB320504the National Hi-Tech Research and Development 863 Program of China under Grant Nos.2008AA01A324 and 2009AA01Z210.
文摘Packet classification on multi-fields is a fundamental mechanism in network equipments,and various classification solutions have been proposed.Because of inherent difficulties,many of these solutions scale poorly in either time or space as rule sets grow in size.Recursive Flow Classification（RFC） is an algorithm with a very high classifying speed. However,its preprocessing complexity and memory requirement are rather high.In this paper,we propose an enhanced RFC（ERFC） algorithm,in which a hash-based aggregated bit vector scheme is exploited to speed up its preprocessing procedure.A compressed and cacheable data structure is also introduced to decrease total memory requirement and improve its searching performance.Evaluation results show that ERFC provides a great improvement over RFC in both space requirement and preprocessing time.The search time complexity of ERFC is equivalent to that of RFC in the worst case; and its average classifying speed is improved by about 100%.
文摘According to the X/Y flow classification method based on TCP and UDP port, a new method named self adaptive X/Y flow classification method is proposed in the paper, which can make the curve of the ratio of label resource usage more stable than ever so as to improve the performance of both L3 forwarding and L2 label switching of LER in MPLS networks. With the simulation of real Internet data, a satisfactory classification result has been obtained.
基金jointly funded by the National Natural Science Foundation of China(grants No.41172104,41202078 and 41372117)the Major National S&T Program of China(grant No.2011ZX05009-002)
文摘Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous debris flows showing complicated sediment composition and sedimentary processes were poorly understood. The main objective of this work is to establish a classification scheme and facies sequence models of subaqueous debris flows for well understanding their sedimentary processes and depositional characteristics.
