Per-flow Re-sequencing in Load-Balanced Switches by Using Dynamic Mailbox Sharing

Size: px
Start display at page:

Download "Per-flow Re-sequencing in Load-Balanced Switches by Using Dynamic Mailbox Sharing"

Transcription

1 Per-flow Re-sequencing in Load-Balanced Switches by Using Dynamic Mailbox Sharing Hong Cheng, Yaohui Jin, Yu Gao, YingDi Yu, Weisheng Hu State Key Laboratory on Fiber-Optic Local Area Networks and Advanced Optical Communication System Shanghai Jiao Tong University, Shanghai , P. R. Nirwan Ansari Advanced Networking Laboratory, NJIT Newark, NJ 07012, USA Abstract Load-balanced switches have received much attention because they are more scalable than other switch architectures. However, a load-balanced switch has the problem of packet missequencing. In this paper, we propose a Dynamic Mailbox Sharing (DMS) scheme to eliminate the mis-sequencing problem of load-balanced switches only at the cost of a very small increase of delay. The key idea is to keep packets of the same flow in order in the load-balanced switch. The DMS scheme is based on two statistical facts in operational networks: the number of simultaneous active flows in the router buffer is far less than that of in-progress flows, and most of the intra-flow packet intervals are longer than the packet delay in the high speed router. In DMS, the packet sequence of the same flow arrived in the input ports is recorded in the mailbox maintained in the output ports. Then, packets of the same flow are delivered according to the order of their arrivals. The mailbox becomes the bottleneck in order to accommodate a large number of flows. We thus propose a dynamic sharing scheme to alleviate the bottleneck and greatly enhance the scalability of the mailbox. By simulations using the real internet traffic traces, we show that with a simple flow splitter mechanism restraining mis-sequencing, the average packet delay using DMS is considerably lower than that of other schemes including Uniform Frame Spreading, Padded Frame and the CR switch, and it is close to the ideal case without resequencing even when the load is very high. The results also demonstrate that the size of mailbox is in the hundreds. Keywords- load-balanced switch; mailbox; dynamic sharing I. INTRODUCTION The load-balanced switch architecture [1], [2] has received much attention because it is considered more scalable than other switch architectures. The basic load-balanced switch consists of a two-stage crossbar. The first stage is used for load balancing. It distributes packets from the input port uniformly to the mediate buffer. The second stage is used for switching, sending packets to their destined output ports. The connection patterns of both stages are deterministic and periodic. So there is no need to use a central scheduler, thus making it more scalable than other switch architectures. The problem of the load-balanced switch architecture is that there are multiple paths between each input/output pair. So packets belonging to the same input/output pair may be delivered out of order [2]. In general, routers shall not deliver packets belonging to the same application flow out of order. Many approaches have been proposed to solve the missequencing problem in load-balanced switches. These approaches can be categorized into two classes. In the first class, a re-sequencing buffer at the output port is used. For example, in the Earliest Deadline First (EDF) scheme, a flow splitter and a load-balancing buffer is added in the input of the load balancing stage to limit the re-sequencing buffer size [2]. However, this scheme requires complicated hardware implementation and non-scalable computation overhead [10]. In the second class, mis-sequencing is prevented throughout both stages and no re-sequencing buffer is needed at the output port. Compared with the first class, the schemes of the second class are more scalable, such as the mailbox switch [11] and Uniform Frame Spreading (UFS) [4] [6]. Although the UFS scheme works well under heavy load, it suffers from the starvation problem under light load because it incurs a large waiting time for accumulating a frame. To improve the performance of UFS under light load, Padded Frame (PF) [3] and CR switch [6] schemes have been proposed. We observed that in the schemes mentioned above, like UFS, PF and CR, not only the order of packets of the same application flow are kept, packets belonging to the same input/output pairs are also delivered in order. (In the following, we will use flow to denote application flow for simplicity.) Yet, routers are not required to deliver packets destined to the same output port in order. That is, this is an unnecessary selfimposed constraint, resulting in redundant re-sequencing operations. The time wasted in redundant re-sequencing unnecessarily increases the delay of packets. Our idea is to just keep packets of the same flow in order in the load-balanced switch. The idea is based on two statistical facts in operational networks. The first one is about the number of flows. Early measurement results show that the number of in-progress flows can be extremely large, rendering per-flow queuing infeasible. However, it has been recently observed that the number of active flows in a switch is measured typically in the hundreds even though there may be tens of thousands of flows in progress [8]. In [5], the authors counted the number of concurrent flows from internet traces. They set the statistical time scale to 10ms to observe the actual number of active flows. They also found that the number of active flows is in the /08/$ IEEE

2 hundreds. Based on this statistical fact, Hu et al. [5] proposed a Dynamic Queue Sharing scheme to implement scalable perflow queuing in high speed routers. The second one is about the intra-flow packet interval. In [7], from the analysis of internet traffic traces, the authors observed that most of the intra-flow packet intervals are more than tens of microseconds, while a packet is typically buffered in a high speed router only for a few microseconds, implying that the probability for a packet encountering mis-sequencing is small. It is expected that the overhead of keeping packets of a flow in order in a load-balanced switch is low. Based on the above two statistical facts, we propose a Dynamic Mailbox Sharing scheme (DMS) to implement perflow re-sequencing in a load-balanced switch. Dynamic mailboxes are put at the output ports of a load balanced switch. The arrival sequence of packets from the same flow is recorded in the mailbox. Then, packets are delivered in accordance with the recorded sequence in the mailbox. To make the implementation of the mailbox scalable, bins in the mailbox are dynamically shared among active flows. When a flow becomes inactive, the bin that the flow occupies is released and can be reused by another active flow. By this dynamic sharing mechanism, the number of bins in the mailbox can be reduced from millions to hundreds. By investigating the distribution the intra-flow packet interval and the delay of packets in the load-balanced switch, we show that most packets have an intra-flow packet interval that leads to a small probability of encountering missequencing, which implies that we can get a much more efficient scheme by avoiding unnecessary maintaining sequence operations. By simulations using real internet traces, we demonstrate that: with a simple flow splitter mechanism restraining mis-sequencing in the input buffer of the first stage, the average delay of packet using DMS outperforms other schemes including Uniform Frame Spreading [4], Padded Frame [3] and the CR switch [6] and it is close to that of the ideal case without re-sequencing. We also demonstrate that the size of mailbox is in the hundreds. The rest of the paper is organized as follows. In Section II, we propose the architecture and operation of DMS in a loadbalanced switch. In Section III, we analyze the probability of packet encountering mis-sequencing. In Section IV, we investigate the performance of the Dynamic Mailbox Sharing, and conduct comparison evaluation with other schemes. Finally, concluding remarks are given in Section V. II. THE SWITCH ARCHITECTURE A. Switch Architecture Fig. 1 shows the architecture of an N N load-balanced switch with Dynamic Mailbox Sharing. As in the basic loadbalanced switch, the switch consists of a two-stage crossbar. In the input buffer of the first stage crossbar, FIFO is used. In the mediate buffer in the input of the second stage crossbar, VOQs are maintained for output ports. In the switch, time is slotted, and cells are of the same size. Figure 1. The architecture of a load-balanced switch with DMS The key components of DMS are the dynamic mailboxes maintained in the output ports. The mailbox is a re-sequencing buffer that can keep the order of packets of the same flow. Each mailbox consists of a number of bins. A bin is allocated to each active flow in the switch. So, in the mailbox, the number of bins in use is not fixed, and thus it is referred to as the dynamic mailbox. The cells in the bin are used to store packets belonging to that flow, and they can also be registered in advance by packets. Since no mechanism is employed in the input buffer to prevent mis-sequencing, packets belonging to the same flow may arrive in the output port out of order. To keep them in order, the sequences of their arrivals in the input ports of the first stage crossbar are recorded by the mailbox through a registration mechanism. Then, the packets arrived in the output port of the second stage crossbar are delivered according to their arrival sequence. We shall next describe how the mailbox works to keep packets of the same flow in order. Registration: When a packet enters the switch, a registration message of the packet is passed to the dynamic mailbox of its destined output port. The registration message includes the ID of the packet and the flow it belongs to. Then the mailbox searches for the bin corresponding to the flow in the message. If no bin is found, a free bin is allocated to the flow. Then, the first empty cell of the bin is allocated to the packet. To record the registration, the cell is tagged with the ID of the packet Sending Packets: In the output port, packets should be sent out in accordance with the sequence in the mailbox registered when they first arrive in the switch. If a packet arrives in the output port earlier than its registered sequence, it is buffered in the mailbox until all the packets registered earlier have been delivered. After a packet has been delivered, its registered cell is released and other registered cells behind are moved forward by one position. B. Dynamic Mailbox Sharing If each in-progress flow is kept in the mailbox, the size of bins in the mailbox can be exorbitantly large. Based on the interesting fact mentioned in the introduction section the number of simultaneous active flows is in the hundreds in a high speed router, we propose a scalable Dynamic Mailbox Sharing mechanism to implement the mailbox that is similar to the previous work of Dynamic Queue Sharing (DQS) [5].

3 Only limited bins are kept in the mailbox and they are shared by the active flows. The mapping between bins and active flows is kept in an Active Flow Mapping (AFM) table (see Fig. 2). When a registration message of a packet from flow fi arrives, the flow identification defined by 5-tuples (source IP address, destination IP address, protocol, source port, and destination port) is searched in the AFM table. If no matched entry is found, a free bin is allocated to this new active flow and a mapping entry is added in the AFM table. When a bin in the mailbox becomes empty that none of its cells is registered, it is released and its mapping entry is deleted from the AFM table. By this dynamic sharing mechanism, the number of bins in the mailbox can be kept in the hundreds, rather than the exorbitant number. By using hashing to further expedite the operations of the search, operations in the AFM table can be performed in only several slots [5]. describe the distribution of intra-flow packet interval, as shown in the first column of TABLE I. Cumulative Probability NLANR-144M CERNET-33M NLANR-66M NLANR-17M Intra-flow Packet Interval / us Figure 3. The Distribution of Intra-flow Packets Interval Figure 2. Using Hashing to Divide the AFM Table to Sub-tables At the end of the section, it is worth to mention the effect of packet segmentation in the load-balanced switch. As we known, various length packets are divided into fixed size cells after entering the switch and these cells are reassembled into complete packet before leaving the switch. Generally it won t bring benefit to the delay of packet directly to keep the cells of a packet in order because a packet can't leave the switch until all of its segments have been assembled, no matter if the segments arrive in order. One inconvenience of out-of-order cells is that we need to use a random accessed queue to reassemble them. In the following sections, we will focus in the problem of packet mis-sequencing. III. THE PROBABILITY OF MIS-SEQUENCING In this section, we investigate the distribution of the intraflow packet interval and how it affects the probability of packet mis-sequencing. As mentioned in the introduction section, the distribution of intra-flow packet intervals can greatly affect mis-sequencing. To see how the distribution is, we analyze some internet packet traces derived from an OC-48 link collected by NLANR [14] as well as collected from CERNET (China Education & Research Network) by DragonLab [15]. We adopt the notation in naming the traces by their source and their average throughput. Fig. 3 shows the distribution of intra-flow packet intervals of various traces. Generally we can see that most of intra-flow packet intervals are larger than tens of microseconds. By summarizing the statistics, we get a group of proportion approximation for different intra-flow packet interval ranges to Another factor that can affect mis-sequencing is the delay of packet in the router. First, let s get a general idea of the distribution of the delay of packet by reviewing the result on measuring the delay of packet in the router in the operational network in previous literatures [12][13]. In [13], the authors have reported the result of measurement of the single-hop packet delay through operational routers in the Spring Internet protocol backbone network. In the measurement, the link utilization is moderate (less than 70%). The result shows that on an OC-12 link, the average queuing delay is only a few microseconds. More than 30% of the packets go through the switch without any queuing delay. The 99th percentile delays are below 50 µs. By comparing with the statistic of intra-flow packet interval, we can see that most packets have an intra-flow packet interval larger than the average packet delay. Under the assumption of uniform and admissible input traffic, the delay in the load-balanced switch is same as that in a equivalent single stage switch with fixed service rate for each VOQ [9][1]. Assume the arrival traffic has an exponential inter-arrival time with mean arrival rate λ and each packet is of fixed one cell size with service rate µ, we can model the packet delay of each input/output pair by the waiting time in an M/D/1 queue, and further approximate that by the waiting time in an M/M/1 queue, with arrival rate (λ/n) and service rate (µ/n). Then the packet delay has a exponential distribution with parameter (µ/n)(1- λ/µ). Take a reasonable implementation of load-balanced switch with N=16, output port rate = OC-192 link speed, cell size = 100 bytes for example. The distribution of packet delay under different load is shown in Fig.4. Next, we shall show that the above properties of intra-flow packet interval and packet delay in the load-balanced switch leads to a small probability of packets being transmitted out-of-order. Assuming packet P1 is blocked in the mailbox because of mis-sequencing, there must be an older packet P2 which belongs to the same flow buffered in the switch. The interval between their arrivals is I. To result in mis-sequencing, the delay time of P2 (denoted as d2) should be larger than the sum of the delay time of P1 (denoted as d1) and I. Only considering

4 the queueing delay portion, we can get an approximation for the probability of P1 encountering mis-sequencing: P d d > I = P d d > I < P d > I (1) ( ) ( q q ) ( q ) By synthesizing all the above results, for the above loadbalanced switch implementation example, we can get the following results about the relationship between the the probability of packet mis-sequencing and intra-flow packet interval under different load, as shown in TABLE II. From TABLE I and TABLE II, we can see that most packets have an intra-flow packet interval which falls into the range that leads to a small probability of packets encountering mis-sequencing, especially when the load is not high. This implies that for a large portion of packets, it is unnecessary to keep their sequence in the price of increasing the packet delay. By using the DMS mechanism, only packets that need to be reordered could be queued in the mailbox. We will count the percentage of packets that have been queued in the mailbox to verify the conclusion of this section by simulations in the following section. Cumulative Probability TABLE I x x x x x10 2 Packet Delay/us load=0.7 load=0.8 load=0.9 load=0.95 Figure 4. Cumulative Probability of Packet Delay THE MIS-SEQUENCING PROBABILITY IN AN LOAD-BLANCED SWITCH IMPLEMENTATION EXAMPLE Intra-packet Interval Proportion Mis-sequencing Probability Approximation Range Load=0.7 Load=0.8 Load=0.9 Load=0.95 [100us, E-4 2.8E-2 [50us, 100us E-5 8E-4 2.8E-2 1.7E-1 [20us, 50us E-2 5.7E-2 2.4E-1 4.9E-1 [10us, 20us E-1 2.4E-1 4.9E-1 7.0E-1 [5us, E-1 4.9E-1 7.0E-1 8.4E-1 IV. SIMULATION In this section, we study the performance of the loadbalanced switch with DMS by simulations using real internet traffic traces. We use the traces from NLANR and CERNET. Since the trace with higher average throughput has smaller average intraflow packet interval, we choose the ones with almost the highest throughput among traces from the same collecting source, thus representing the worst case scenario. We inject the traffic of various traces as input into the proposed loadbalanced switch. Each input port starts reading the trace file from different location. Flows in traces are mapped to output ports by ensuring that the average packet arrival rate to each output is the same and does not exceed the output port capacity. In the simulations, the load-balanced switch is time slotted, and processes fixed-size cells. Since packets from the traces are of variable length, they are segmented to 64 bytes cells before entering the switch. To adjust the arrival rate, the length of a slot is changed according to the throughput of the trace. For example, if the throughput of a trace is 500 Mb/s, to achieve the arrival rate of 0.8, the slot (a time unit to transmit a 64-byte cell) length is ((64 8)/500M) 0.8 = µs. Each experiment lasts at least 106 slots A. Average Delay We first study the average delay of the proposed loadbalanced switch with DMS, and then compare its performance in terms of the average delay with that of other schemes including UFS, PF and the CR switch. (Because the mailbox switch [11] is similar to the CR switch [6] working in the contention mode, we do not include it in the comparison.) To show the extra delay caused by re-sequencing clearly, we also plot the average delay of the basic load-balanced switch (BLBS) without any re-sequencing process. The results of applying CERNET-33M traces to various switches are shown in Fig. 5. (For page limitation, we do not show the figure of the result of simulations with NLANR-133M traces.) The results for CR, UFS and PF using the CERNET-33M trace (in Fig. 5) conform to those reported in [3] and [6] when the input traffic is set to be bursty. We can observe from both figures that when the load is not high ([0, 0.8] in Fig.5), the performance of DMS outperforms others (PF, CR, and UFS) and is very close to that of BLBS (the ideal case). By avoiding unnecessary maintaining sequence operations, DMS keeps packets of the same flow in order and only makes a little sacrifice on the packet delay. However, when the load is extremely high, the average delay of DMS is not as well as other schemes. To improve the performance of DMS in the high load case, we can use some mechanism that is able to restrain mis-sequencing in the input buffer of first stage. Chang et al. [2] proposed to maintain a flow splitter and a load-balancing buffer in the input buffer. The load-balancing buffer in the input of first stage consists of N VOQs destined for the N output ports of first stage. Packets belonging to the same input/output pair are split in the roundrobin fashion to the mediate buffer so that the lengths of VOQs in mediate buffers are approximately the same and the delay of re-sequencing can be reduced. We compare the average delay of DMS with flow splitter with other schemes (UFS, PF, and CR) in the high load case by simulations in a switch with the CERNET-33M trace. We observe a similar result as that in the mediate load case: the average delay of DMS is very close to BLBS (the ideal case) and it outperforms other schemes including UFS, PF and CR in the high load case. We also observe that the difference

5 between the average delay of DMS with flow splitter and the ideal case doesn t increase with the load increasing. Average Delay BLBS DMS UFS PF CR Arrival Rate Figure 5. The Average Delay Using the CERNET-33M Trace B. Mailbox In this section, we investigate the details of the working of mailbox. We examine the percentage of packets that have been queued in the mailbox in the simulations of the switch using the CERNET-33M trace, as shown in TABLE II. We also show the number of bins in use and the number of registered cells in the table. From TABLE II, we observe that the percentage of packets queued in mailbox is small. Even when the load is as high as 0.9, the percentage is less than 0.1. So it is fairly unnecessary to perform keeping sequence operations for every packet. We also observe that on average there are only several registered cells in the bin of mailbox, which represents the average number of packets of an active flow in the switch. The maximum number of bins in use is in the low hundreds. Hence, the size of the mailbox (the number of bins) can be set to only several hundreds which are sufficient to keep all the active flows. V. CONCLUSION In this paper, we have proposed a Dynamic Mailbox Sharing mechanism for load-balanced switches to solve the mis-sequencing problem. In the DMS mechanism, the sequence of arrivals of packets of the same flow is recorded in the mailbox maintained in the output port. Packets of the same flow are delivered according to their arrival sequence recorded in the mailbox. To achieve scalability, bins in the mailbox are dynamically shared among active flows. By the dynamic sharing mechanism, the number of bins in the mailbox can be kept in the low hundreds. By simulations, we have also demonstrated that with a simple flow splitter mechanism in the input buffer, the average delay of DMS is considerable smaller than that of other schemes including Uniform Frame Spreading, Padded Frame, and CR switches. The effect of packet segmentation requires further research and will be part of our future endeavor. Arrival Rate Maximum Number of Bins in use TABLE II. Average Number of Bins in use STATISTICS IN MAILBOX Maximum Number Average Number of Registered Cells of Registered Cells in the Bin in the Bin Percentage of Packets Queued in Mailbox The authors would like to thank Professor Wende Zhong (NTU, Singapore) for his discussion. REFERENCES [1] C.S. Chang, D.S Lee, and Y.S Jou, Load balanced Birkhoff-von Neumann switches, part I: one-stage buffering, Computer Communications, Vol.25, pp , [2] C.S. Chang, D.S Lee, and C.M Lien, Load balanced Birkhoff-von Neumann switches, part II: Multi-stage buffering, Computer Communications, Vol.25, pp , [3] J.J Jaramillo, F. Milan, and R. Srikant, Padded frames: a novel algorithm for stable scheduling in load-balanced switches, Proceedings of CISS, Princeton, NJ, March [4] I. Keslassy, The load-balanced router, Ph.D. dissertation, Stanford University, Stanford, CA, USA, [5] C. Hu, Y. Tang, X. Chen, and B. Liu, Per-flow Queueing by Dynamic Queue Sharing, Proceedings of IEEE INFOCOM, Anchorage, Alaska, [6] C.L. Yu, C.S. Chang, and D.S. Lee, CR Switch: A Load-Balanced Switch with Contention and Reservation, Proceedings of IEEE INFOCOM, Anchorage, Alaska, [7] L. Shi, Y. Zhang, J. Yu, B. Xu, B. Liu, and J. Li, On the Extreme Parallelism Inside Next-Generation Network Processors, Proceedings of IEEE INFOCOM, Anchorage, Alaska, 2007 [8] A. Kortebi, L. Muscariello, S. Oueslati, and J. Roberts, Evaluating the number of active flows in a scheduler realizing fair statistical bandwidth sharing, in ACM SIGMETRICS 2005, 2005, pp [9] I. Keslassy, S.T Chuang, K. Yu, D Miller, M. Horowitz, O. Solgaard, and N. McKeown, Scaling internet routers using optics, Proceedings of ACM SIGCOMM, Karlsruhe, Germany, August [10] C.S. Chang, D.S. Lee, and C.Y. Yue, Providing guaranteed rate services in the load balanced Birkhoff-von Neumann switches, Proceedings of IEEE INFOCOM, 2003 [11] C.-S. Chang, D.-S. Lee and Y.-J. Shih, Mailbox switch: a scalable twostage switch architecture for conflict resolution of ordered packets, Proceedings of IEEE INFOCOM, Vol. 3, pp , Hong Kong, [12] B.Y. Choi, S. Moon, Z.L. Zhang, K. Papagiannaki, C. Diot. "Analysis of Point-To-Point Packet Delay In an Operational Network," Proceedings of INFOCOM, March, 2004, Hong Kong. [13] K. Papagiannaki, S. Moon, C. Fraleigh, P. Thiran, and C. Diot. "Measurement and analysis of single-hop delay on an IP backbone network," Proceedings of INFOCOM, San Francisco, April [14] NLANR. Passive measurement and analysis (pma). [Online]. Available: [15] DRAGON-Lab. CERNET trace download. [Online]. Available:

A QOS BASED LOAD BALANCED SWITCH

A QOS BASED LOAD BALANCED SWITCH A QOS BASED LOAD BALANCED SWITCH Vishnu Prasath Badri narayanan, Manigandan S K Veltech Hightech Dr.RR and Dr.SR Engineering college,india ABSTRACT The simple architecture with high forwarding capacity

More information

Load Balancing and Switch Scheduling

Load Balancing and Switch Scheduling EE384Y Project Final Report Load Balancing and Switch Scheduling Xiangheng Liu Department of Electrical Engineering Stanford University, Stanford CA 94305 Email: liuxh@systems.stanford.edu Abstract Load

More information

Performance Analysis of a Practical Load Balanced Switch

Performance Analysis of a Practical Load Balanced Switch Performance Analysis of a Practical Balanced Switch Yanming Shen, Shivendra S Panwar, H Jonathan Chao Department of Electrical and Computer Engineering Polytechnic University Abstract The load balanced

More information

1212 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 5, OCTOBER 2008

1212 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 5, OCTOBER 2008 1212 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 5, OCTOBER 2008 Padded Frames: A Novel Algorithm for Stable Scheduling in Load-Balanced Switches Juan José Jaramillo, Student Member, IEEE, Fabio

More information

An Asymptotically Minimal Node-degree Topology for Load-Balanced Architectures

An Asymptotically Minimal Node-degree Topology for Load-Balanced Architectures An Asymptotically Minimal ode-degree Topology for Load-Balanced Architectures Zhenhua Liu, Xiaoping Zhang, Youjian Zhao, Hongtao Guan Department of Computer Science and Technology, Tsinghua University

More information

IEEE/ACM TRANSACTIONS ON NETWORKING 1 1063-6692/$26.00 2009 IEEE

IEEE/ACM TRANSACTIONS ON NETWORKING 1 1063-6692/$26.00 2009 IEEE IEEE/ACM TRANSACTIONS ON NETWORKING 1 Feedback-Based Scheduling for Load-Balanced Two-Stage Switches Bing Hu, Student Member, IEEE, and Kwan L. Yeung, Senior Member, IEEE Abstract A framework for designing

More information

MLPPP Deployment Using the PA-MC-T3-EC and PA-MC-2T3-EC

MLPPP Deployment Using the PA-MC-T3-EC and PA-MC-2T3-EC MLPPP Deployment Using the PA-MC-T3-EC and PA-MC-2T3-EC Overview Summary The new enhanced-capability port adapters are targeted to replace the following Cisco port adapters: 1-port T3 Serial Port Adapter

More information

PART III. OPS-based wide area networks

PART III. OPS-based wide area networks PART III OPS-based wide area networks Chapter 7 Introduction to the OPS-based wide area network 7.1 State-of-the-art In this thesis, we consider the general switch architecture with full connectivity

More information

Network Performance Monitoring at Small Time Scales

Network Performance Monitoring at Small Time Scales Network Performance Monitoring at Small Time Scales Konstantina Papagiannaki, Rene Cruz, Christophe Diot Sprint ATL Burlingame, CA dina@sprintlabs.com Electrical and Computer Engineering Department University

More information

Monitoring Large Flows in Network

Monitoring Large Flows in Network Monitoring Large Flows in Network Jing Li, Chengchen Hu, Bin Liu Department of Computer Science and Technology, Tsinghua University Beijing, P. R. China, 100084 { l-j02, hucc03 }@mails.tsinghua.edu.cn,

More information

An Adaptive Load Balancing to Provide Quality of Service

An Adaptive Load Balancing to Provide Quality of Service An Adaptive Load Balancing to Provide Quality of Service 1 Zahra Vali, 2 Massoud Reza Hashemi, 3 Neda Moghim *1, Isfahan University of Technology, Isfahan, Iran 2, Isfahan University of Technology, Isfahan,

More information

Concurrent Round-Robin-Based Dispatching Schemes for Clos-Network Switches

Concurrent Round-Robin-Based Dispatching Schemes for Clos-Network Switches 830 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 10, NO. 6, DECEMBER 2002 Concurrent Round-Robin-Based Dispatching Schemes for Clos-Network Switches Eiji Oki, Member, IEEE, Zhigang Jing, Member, IEEE, Roberto

More information

VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications

VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications Veselin Rakocevic School of Engineering and Mathematical Sciences City University, London, UK V.Rakocevic@city.ac.uk

More information

Chapter 4. VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network)

Chapter 4. VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network) Chapter 4 VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network) 4.1 Introduction Traffic Engineering can be defined as a task of mapping traffic

More information

FPGA Implementation of IP Packet Segmentation and Reassembly in Internet Router*

FPGA Implementation of IP Packet Segmentation and Reassembly in Internet Router* SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol. 6, No. 3, December 2009, 399-407 UDK: 004.738.5.057.4 FPGA Implementation of IP Packet Segmentation and Reassembly in Internet Router* Marko Carević 1,a,

More information

Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation

Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation R.Navaneethakrishnan Assistant Professor (SG) Bharathiyar College of Engineering and Technology, Karaikal, India.

More information

NOVEL PRIORITISED EGPRS MEDIUM ACCESS REGIME FOR REDUCED FILE TRANSFER DELAY DURING CONGESTED PERIODS

NOVEL PRIORITISED EGPRS MEDIUM ACCESS REGIME FOR REDUCED FILE TRANSFER DELAY DURING CONGESTED PERIODS NOVEL PRIORITISED EGPRS MEDIUM ACCESS REGIME FOR REDUCED FILE TRANSFER DELAY DURING CONGESTED PERIODS D. Todinca, P. Perry and J. Murphy Dublin City University, Ireland ABSTRACT The goal of this paper

More information

Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks

Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks Prashil S. Waghmare PG student, Sinhgad College of Engineering, Vadgaon, Pune University, Maharashtra, India. prashil.waghmare14@gmail.com

More information

Comparative Analysis of Congestion Control Algorithms Using ns-2

Comparative Analysis of Congestion Control Algorithms Using ns-2 www.ijcsi.org 89 Comparative Analysis of Congestion Control Algorithms Using ns-2 Sanjeev Patel 1, P. K. Gupta 2, Arjun Garg 3, Prateek Mehrotra 4 and Manish Chhabra 5 1 Deptt. of Computer Sc. & Engg,

More information

Ternary-Search-based Scheme to Measure Link Available-bandwidth in Wired Networks

Ternary-Search-based Scheme to Measure Link Available-bandwidth in Wired Networks Ternary-Search-based Scheme to Measure Link Available-bandwidth in Wired Networks Khondaker M. Salehin and Roberto Rojas-Cessa Networking Research Laboratory Department of Electrical and Computer Engineering

More information

Providing Deterministic Quality-of-Service Guarantees on WDM Optical Networks

Providing Deterministic Quality-of-Service Guarantees on WDM Optical Networks 2072 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 18, NO. 10, OCTOBER 2000 Providing Deterministic Quality-of-Service Guarantees on WDM Optical Networks Maode Ma and Mounir Hamdi, Member, IEEE

More information

Analysis of IP Network for different Quality of Service

Analysis of IP Network for different Quality of Service 2009 International Symposium on Computing, Communication, and Control (ISCCC 2009) Proc.of CSIT vol.1 (2011) (2011) IACSIT Press, Singapore Analysis of IP Network for different Quality of Service Ajith

More information

Traffic Behavior Analysis with Poisson Sampling on High-speed Network 1

Traffic Behavior Analysis with Poisson Sampling on High-speed Network 1 Traffic Behavior Analysis with Poisson Sampling on High-speed etwork Guang Cheng Jian Gong (Computer Department of Southeast University anjing 0096, P.R.China) Abstract: With the subsequent increasing

More information

A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster

A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster , pp.11-20 http://dx.doi.org/10.14257/ ijgdc.2014.7.2.02 A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster Kehe Wu 1, Long Chen 2, Shichao Ye 2 and Yi Li 2 1 Beijing

More information

Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow

Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow International Journal of Soft Computing and Engineering (IJSCE) Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow Abdullah Al Masud, Hossain Md. Shamim, Amina Akhter

More information

A Power Efficient QoS Provisioning Architecture for Wireless Ad Hoc Networks

A Power Efficient QoS Provisioning Architecture for Wireless Ad Hoc Networks A Power Efficient QoS Provisioning Architecture for Wireless Ad Hoc Networks Didem Gozupek 1,Symeon Papavassiliou 2, Nirwan Ansari 1, and Jie Yang 1 1 Department of Electrical and Computer Engineering

More information

Threshold-based Exhaustive Round-robin for the CICQ Switch with Virtual Crosspoint Queues

Threshold-based Exhaustive Round-robin for the CICQ Switch with Virtual Crosspoint Queues Threshold-based Exhaustive Round-robin for the CICQ Switch with Virtual Crosspoint Queues Kenji Yoshigoe Department of Computer Science University of Arkansas at Little Rock Little Rock, AR 7224 kxyoshigoe@ualr.edu

More information

ROUTER CONTROL MECHANISM FOR CONGESTION AVOIDANCE IN CDMA BASED IP NETWORK

ROUTER CONTROL MECHANISM FOR CONGESTION AVOIDANCE IN CDMA BASED IP NETWORK International Journal of Information Technology and Knowledge Management July-December 2010, Volume 2, No. 2, pp. 465-470 ROUTER CONTROL MECHANISM FOR CONGESTION AVOIDANCE IN CDMA BASED IP NETWORK V.Sumalatha1

More information

Optical interconnection networks with time slot routing

Optical interconnection networks with time slot routing Theoretical and Applied Informatics ISSN 896 5 Vol. x 00x, no. x pp. x x Optical interconnection networks with time slot routing IRENEUSZ SZCZEŚNIAK AND ROMAN WYRZYKOWSKI a a Institute of Computer and

More information

Smart Queue Scheduling for QoS Spring 2001 Final Report

Smart Queue Scheduling for QoS Spring 2001 Final Report ENSC 833-3: NETWORK PROTOCOLS AND PERFORMANCE CMPT 885-3: SPECIAL TOPICS: HIGH-PERFORMANCE NETWORKS Smart Queue Scheduling for QoS Spring 2001 Final Report By Haijing Fang(hfanga@sfu.ca) & Liu Tang(llt@sfu.ca)

More information

Load Balancing Mechanisms in Data Center Networks

Load Balancing Mechanisms in Data Center Networks Load Balancing Mechanisms in Data Center Networks Santosh Mahapatra Xin Yuan Department of Computer Science, Florida State University, Tallahassee, FL 33 {mahapatr,xyuan}@cs.fsu.edu Abstract We consider

More information

Transport Layer Protocols

Transport Layer Protocols Transport Layer Protocols Version. Transport layer performs two main tasks for the application layer by using the network layer. It provides end to end communication between two applications, and implements

More information

White Paper Abstract Disclaimer

White Paper Abstract Disclaimer White Paper Synopsis of the Data Streaming Logical Specification (Phase I) Based on: RapidIO Specification Part X: Data Streaming Logical Specification Rev. 1.2, 08/2004 Abstract The Data Streaming specification

More information

Behavior Analysis of TCP Traffic in Mobile Ad Hoc Network using Reactive Routing Protocols

Behavior Analysis of TCP Traffic in Mobile Ad Hoc Network using Reactive Routing Protocols Behavior Analysis of TCP Traffic in Mobile Ad Hoc Network using Reactive Routing Protocols Purvi N. Ramanuj Department of Computer Engineering L.D. College of Engineering Ahmedabad Hiteishi M. Diwanji

More information

Research on Errors of Utilized Bandwidth Measured by NetFlow

Research on Errors of Utilized Bandwidth Measured by NetFlow Research on s of Utilized Bandwidth Measured by NetFlow Haiting Zhu 1, Xiaoguo Zhang 1,2, Wei Ding 1 1 School of Computer Science and Engineering, Southeast University, Nanjing 211189, China 2 Electronic

More information

RESOURCE ALLOCATION FOR INTERACTIVE TRAFFIC CLASS OVER GPRS

RESOURCE ALLOCATION FOR INTERACTIVE TRAFFIC CLASS OVER GPRS RESOURCE ALLOCATION FOR INTERACTIVE TRAFFIC CLASS OVER GPRS Edward Nowicki and John Murphy 1 ABSTRACT The General Packet Radio Service (GPRS) is a new bearer service for GSM that greatly simplify wireless

More information

Performance Analysis of Storage Area Network Switches

Performance Analysis of Storage Area Network Switches Performance Analysis of Storage Area Network Switches Andrea Bianco, Paolo Giaccone, Enrico Maria Giraudo, Fabio Neri, Enrico Schiattarella Dipartimento di Elettronica - Politecnico di Torino - Italy e-mail:

More information

Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT:

Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT: Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT: In view of the fast-growing Internet traffic, this paper propose a distributed traffic management

More information

Computing Load Aware and Long-View Load Balancing for Cluster Storage Systems

Computing Load Aware and Long-View Load Balancing for Cluster Storage Systems 215 IEEE International Conference on Big Data (Big Data) Computing Load Aware and Long-View Load Balancing for Cluster Storage Systems Guoxin Liu and Haiying Shen and Haoyu Wang Department of Electrical

More information

Path Selection Methods for Localized Quality of Service Routing

Path Selection Methods for Localized Quality of Service Routing Path Selection Methods for Localized Quality of Service Routing Xin Yuan and Arif Saifee Department of Computer Science, Florida State University, Tallahassee, FL Abstract Localized Quality of Service

More information

OpenFlow Based Load Balancing

OpenFlow Based Load Balancing OpenFlow Based Load Balancing Hardeep Uppal and Dane Brandon University of Washington CSE561: Networking Project Report Abstract: In today s high-traffic internet, it is often desirable to have multiple

More information

ECE 358: Computer Networks. Solutions to Homework #4. Chapter 4 - The Network Layer

ECE 358: Computer Networks. Solutions to Homework #4. Chapter 4 - The Network Layer ECE 358: Computer Networks Solutions to Homework #4 Chapter 4 - The Network Layer P 4. Consider the network below. a. Suppose that this network is a datagram network. Show the forwarding table in router

More information

PFS scheme for forcing better service in best effort IP network

PFS scheme for forcing better service in best effort IP network Paper PFS scheme for forcing better service in best effort IP network Monika Fudała and Wojciech Burakowski Abstract The paper presents recent results corresponding to a new strategy for source traffic

More information

Maximizing the number of users in an interactive video-ondemand. Citation Ieee Transactions On Broadcasting, 2002, v. 48 n. 4, p.

Maximizing the number of users in an interactive video-ondemand. Citation Ieee Transactions On Broadcasting, 2002, v. 48 n. 4, p. Title Maximizing the number of users in an interactive video-ondemand system Author(s) Bakiras, S; Li, VOK Citation Ieee Transactions On Broadcasting, 2002, v. 48 n. 4, p. 281-292 Issued Date 2002 URL

More information

Making Money With Affiliate Marketing - How to Do It

Making Money With Affiliate Marketing - How to Do It A New Dynamic Bandwidth Allocation Protocol with Quality of Service in Ethernetbased Passive Optical Networks Fu-Tai An, Yu-Li Hsueh, Kyeong Soo Kim, Ian M White, and Leonid G Kazovsky Optical Communication

More information

Adaptive Batch Scheduling for Packet Switching with Delays

Adaptive Batch Scheduling for Packet Switching with Delays 1 Adaptive Batch Scheduling for Packet Switching with Delays Kevin Ross and Nicholas Bambos 1 University of California Santa Cruz, kross@soe.usc.edu Stanford University, bambos@stanford.edu We discuss

More information

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING CHAPTER 6 CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING 6.1 INTRODUCTION The technical challenges in WMNs are load balancing, optimal routing, fairness, network auto-configuration and mobility

More information

TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) Internet Protocol (IP)

TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) Internet Protocol (IP) TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) *Slides adapted from a talk given by Nitin Vaidya. Wireless Computing and Network Systems Page

More information

Network Layer: Network Layer and IP Protocol

Network Layer: Network Layer and IP Protocol 1 Network Layer: Network Layer and IP Protocol Required reading: Garcia 7.3.3, 8.1, 8.2.1 CSE 3213, Winter 2010 Instructor: N. Vlajic 2 1. Introduction 2. Router Architecture 3. Network Layer Protocols

More information

Final for ECE374 05/06/13 Solution!!

Final for ECE374 05/06/13 Solution!! 1 Final for ECE374 05/06/13 Solution!! Instructions: Put your name and student number on each sheet of paper! The exam is closed book. You have 90 minutes to complete the exam. Be a smart exam taker -

More information

4 Internet QoS Management

4 Internet QoS Management 4 Internet QoS Management Rolf Stadler School of Electrical Engineering KTH Royal Institute of Technology stadler@ee.kth.se September 2008 Overview Network Management Performance Mgt QoS Mgt Resource Control

More information

Optical Packet Buffers with Active Queue Management

Optical Packet Buffers with Active Queue Management Optical Packet Buffers with Active Queue Management Assaf Shacham and Keren Bergman Columbia University, Department of Electrical Engineering, New York, NY 10027 bergman@ee.columbia.edu, http://lightwave.ee.columbia.edu

More information

Radio Resource Allocation in GSM/GPRS Networks

Radio Resource Allocation in GSM/GPRS Networks Radio Resource Allocation in GSM/GPRS Networks Jean-Lien C. Wu 1, Wei-Yeh Chen 2, and Hung-Huan Liu 1 1 Department of Electronic Engineering, National Taiwan University of Science and Technology, 43, Keelung

More information

packet retransmitting based on dynamic route table technology, as shown in fig. 2 and 3.

packet retransmitting based on dynamic route table technology, as shown in fig. 2 and 3. Implementation of an Emulation Environment for Large Scale Network Security Experiments Cui Yimin, Liu Li, Jin Qi, Kuang Xiaohui National Key Laboratory of Science and Technology on Information System

More information

Bandwidth Allocation DBA (BA-DBA) Algorithm for xpon Networks

Bandwidth Allocation DBA (BA-DBA) Algorithm for xpon Networks Bandwidth Allocation DBA (BA-DBA) Algorithm for xpon Networks Karamitsos Ioannis Telecom Lab,On Telecoms Athens, Attica 16342, Greece Al-Arfaj Khalid Rayadah Investment Co Riyadh, KSA Kingdom of Saudi

More information

A Review on Quality of Service Architectures for Internet Network Service Provider (INSP)

A Review on Quality of Service Architectures for Internet Network Service Provider (INSP) A Review on Quality of Service Architectures for Internet Network Service Provider (INSP) Herman and Azizah bte Abd. Rahman Faculty of Computer Science and Information System Universiti Teknologi Malaysia

More information

Scheduling for QoS Management

Scheduling for QoS Management Scheduling for QoS Management Domenico Massimo Parrucci Condello isti information science Facoltà and di Scienze technology e Tecnologie institute 1/number 1 Outline What is Queue Management and Scheduling?

More information

A Content-Based Load Balancing Algorithm for Metadata Servers in Cluster File Systems*

A Content-Based Load Balancing Algorithm for Metadata Servers in Cluster File Systems* A Content-Based Load Balancing Algorithm for Metadata Servers in Cluster File Systems* Junho Jang, Saeyoung Han, Sungyong Park, and Jihoon Yang Department of Computer Science and Interdisciplinary Program

More information

Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration

Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration 1 Harish H G, 2 Dr. R Girisha 1 PG Student, 2 Professor, Department of CSE, PESCE Mandya (An Autonomous Institution under

More information

Network congestion control using NetFlow

Network congestion control using NetFlow Network congestion control using NetFlow Maxim A. Kolosovskiy Elena N. Kryuchkova Altai State Technical University, Russia Abstract The goal of congestion control is to avoid congestion in network elements.

More information

TCP and UDP Performance for Internet over Optical Packet-Switched Networks

TCP and UDP Performance for Internet over Optical Packet-Switched Networks TCP and UDP Performance for Internet over Optical Packet-Switched Networks Jingyi He S-H Gary Chan Department of Electrical and Electronic Engineering Department of Computer Science Hong Kong University

More information

Design and Performance Analysis of a Practical Load-Balanced Switch

Design and Performance Analysis of a Practical Load-Balanced Switch 242 IEEE TRANSACTIONS ON COMMUNICATIONS, VOL 57, NO 8, AUGUST 29 Design and Performance Analysis of a Practical Load-Balanced Switch Yanming Shen, Shivendra S Panwar, and H Jonathan Chao Abstract The load-balanced

More information

Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics

Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics Proceedings of the 4 IEEE United States Military Academy, West Point, NY - June Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics Wei Yan, Student Member, IEEE, and Edwin

More information

Algorithms for Interference Sensing in Optical CDMA Networks

Algorithms for Interference Sensing in Optical CDMA Networks Algorithms for Interference Sensing in Optical CDMA Networks Purushotham Kamath, Joseph D. Touch and Joseph A. Bannister {pkamath, touch, joseph}@isi.edu Information Sciences Institute, University of Southern

More information

Performance advantages of resource sharing in polymorphic optical networks

Performance advantages of resource sharing in polymorphic optical networks R. J. Durán, I. de Miguel, N. Merayo, P. Fernández, R. M. Lorenzo, E. J. Abril, I. Tafur Monroy, Performance advantages of resource sharing in polymorphic optical networks, Proc. of the 0th European Conference

More information

Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks

Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks Kam-Yiu Lam 1, Joe Yuen 1, Sang H. Son 2 and Edward Chan 1 Department of Computer Science 1 City University

More information

A TCP-like Adaptive Contention Window Scheme for WLAN

A TCP-like Adaptive Contention Window Scheme for WLAN A TCP-like Adaptive Contention Window Scheme for WLAN Qixiang Pang, Soung Chang Liew, Jack Y. B. Lee, Department of Information Engineering The Chinese University of Hong Kong Hong Kong S.-H. Gary Chan

More information

A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11

A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11 A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11 Christian Röhl, Hagen Woesner, Adam Wolisz * Technical University Berlin Telecommunication Networks Group {roehl,

More information

Assessment of Traffic Prioritization in Switched Local Area Networks Carrying Multimedia Traffic

Assessment of Traffic Prioritization in Switched Local Area Networks Carrying Multimedia Traffic Assessment of Traffic Prioritization in Switched Local Area Networks Carrying Multimedia Traffic F. A. Tobagi, C. J. Fraleigh, M. J. Karam, W. Noureddine Computer Systems Laboratory Department of Electrical

More information

Module 5. Broadcast Communication Networks. Version 2 CSE IIT, Kharagpur

Module 5. Broadcast Communication Networks. Version 2 CSE IIT, Kharagpur Module 5 Broadcast Communication Networks Lesson 1 Network Topology Specific Instructional Objectives At the end of this lesson, the students will be able to: Specify what is meant by network topology

More information

Task Scheduling in Hadoop

Task Scheduling in Hadoop Task Scheduling in Hadoop Sagar Mamdapure Munira Ginwala Neha Papat SAE,Kondhwa SAE,Kondhwa SAE,Kondhwa Abstract Hadoop is widely used for storing large datasets and processing them efficiently under distributed

More information

Configuring a Load-Balancing Scheme

Configuring a Load-Balancing Scheme Configuring a Load-Balancing Scheme Last Updated: October 5, 2011 This module contains information about Cisco Express Forwarding and describes the tasks for configuring a load-balancing scheme for Cisco

More information

COMPARATIVE ANALYSIS OF DIFFERENT QUEUING MECHANISMS IN HETROGENEOUS NETWORKS

COMPARATIVE ANALYSIS OF DIFFERENT QUEUING MECHANISMS IN HETROGENEOUS NETWORKS COMPARATIVE ANALYSIS OF DIFFERENT QUEUING MECHANISMS IN HETROGENEOUS NETWORKS Shubhangi Rastogi 1, Samir Srivastava 2 M.Tech Student, Computer Science and Engineering, KNIT, Sultanpur, India 1 Associate

More information

Using median filtering in active queue management for telecommunication networks

Using median filtering in active queue management for telecommunication networks Using median filtering in active queue management for telecommunication networks Sorin ZOICAN *, Ph.D. Cuvinte cheie. Managementul cozilor de aşteptare, filtru median, probabilitate de rejectare, întârziere.

More information

Rohde & Schwarz R&S SITLine ETH VLAN Encryption Device Functionality & Performance Tests

Rohde & Schwarz R&S SITLine ETH VLAN Encryption Device Functionality & Performance Tests Rohde & Schwarz R&S Encryption Device Functionality & Performance Tests Introduction Following to our test of the Rohde & Schwarz ETH encryption device in April 28 the European Advanced Networking Test

More information

Cloud Storage and Online Bin Packing

Cloud Storage and Online Bin Packing Cloud Storage and Online Bin Packing Doina Bein, Wolfgang Bein, and Swathi Venigella Abstract We study the problem of allocating memory of servers in a data center based on online requests for storage.

More information

A Dynamic Link Allocation Router

A Dynamic Link Allocation Router A Dynamic Link Allocation Router Wei Song and Doug Edwards School of Computer Science, the University of Manchester Oxford Road, Manchester M13 9PL, UK {songw, doug}@cs.man.ac.uk Abstract The connection

More information

SUPPORT FOR HIGH-PRIORITY TRAFFIC IN VLSI COMMUNICATION SWITCHES

SUPPORT FOR HIGH-PRIORITY TRAFFIC IN VLSI COMMUNICATION SWITCHES 9th Real-Time Systems Symposium Huntsville, Alabama, pp 191-, December 1988 SUPPORT FOR HIGH-PRIORITY TRAFFIC IN VLSI COMMUNICATION SWITCHES Yuval Tamir and Gregory L Frazier Computer Science Department

More information

Edge-based active queue management

Edge-based active queue management Edge-based active queue management L. Zhu, N. Ansari, G. Cheng and K. Xu Abstract: In the paper, a new framework of active queue management, namely, edge-based active queue management (), is proposed.

More information

Design and Implementation of an On-Chip timing based Permutation Network for Multiprocessor system on Chip

Design and Implementation of an On-Chip timing based Permutation Network for Multiprocessor system on Chip Design and Implementation of an On-Chip timing based Permutation Network for Multiprocessor system on Chip Ms Lavanya Thunuguntla 1, Saritha Sapa 2 1 Associate Professor, Department of ECE, HITAM, Telangana

More information

Designing a Predictable Internet Backbone with Valiant Load-Balancing

Designing a Predictable Internet Backbone with Valiant Load-Balancing Designing a Predictable Internet Backbone with Valiant Load-Balancing ui Zhang-Shen and ick McKeown Computer Systems Laboratory, Stanford University, Stanford, CA 94305-9030, USA {rzhang, nickm}@stanford.edu

More information

Load Balancing on a Grid Using Data Characteristics

Load Balancing on a Grid Using Data Characteristics Load Balancing on a Grid Using Data Characteristics Jonathan White and Dale R. Thompson Computer Science and Computer Engineering Department University of Arkansas Fayetteville, AR 72701, USA {jlw09, drt}@uark.edu

More information

An Active Network Based Hierarchical Mobile Internet Protocol Version 6 Framework

An Active Network Based Hierarchical Mobile Internet Protocol Version 6 Framework An Active Network Based Hierarchical Mobile Internet Protocol Version 6 Framework Zutao Zhu Zhenjun Li YunYong Duan Department of Business Support Department of Computer Science Department of Business

More information

Basic Multiplexing models. Computer Networks - Vassilis Tsaoussidis

Basic Multiplexing models. Computer Networks - Vassilis Tsaoussidis Basic Multiplexing models? Supermarket?? Computer Networks - Vassilis Tsaoussidis Schedule Where does statistical multiplexing differ from TDM and FDM Why are buffers necessary - what is their tradeoff,

More information

Randomized Load Balancing and Oblivious Routing

Randomized Load Balancing and Oblivious Routing Randomized Load Balancing and Oblivious Routing Peter J. Winzer Bell Labs, Alcatel-Lucent Joint work with F. B. Shepherd, M. K. Thottan, S. Borst, R. Prasad DIMACS Tutorial on Algorithms for Next Generation

More information

Performance Modeling and Analysis of a Database Server with Write-Heavy Workload

Performance Modeling and Analysis of a Database Server with Write-Heavy Workload Performance Modeling and Analysis of a Database Server with Write-Heavy Workload Manfred Dellkrantz, Maria Kihl 2, and Anders Robertsson Department of Automatic Control, Lund University 2 Department of

More information

A Catechistic Method for Traffic Pattern Discovery in MANET

A Catechistic Method for Traffic Pattern Discovery in MANET A Catechistic Method for Traffic Pattern Discovery in MANET R. Saranya 1, R. Santhosh 2 1 PG Scholar, Computer Science and Engineering, Karpagam University, Coimbatore. 2 Assistant Professor, Computer

More information

Entropy-Based Collaborative Detection of DDoS Attacks on Community Networks

Entropy-Based Collaborative Detection of DDoS Attacks on Community Networks Entropy-Based Collaborative Detection of DDoS Attacks on Community Networks Krishnamoorthy.D 1, Dr.S.Thirunirai Senthil, Ph.D 2 1 PG student of M.Tech Computer Science and Engineering, PRIST University,

More information

Sprinklers: A Randomized Variable-Size Striping Approach to Reordering-Free Load-Balanced Switching

Sprinklers: A Randomized Variable-Size Striping Approach to Reordering-Free Load-Balanced Switching Sprinklers: A Randomized Variable-Size Striping Approach to Reordering-Free Load-Balanced Switching Weijun Ding Jim Xu Jim Dai Yang Song Bill Lin June 7, 2014 Abstract Internet traffic continues to grow

More information

Path Selection Analysis in MPLS Network Based on QoS

Path Selection Analysis in MPLS Network Based on QoS Cumhuriyet Üniversitesi Fen Fakültesi Fen Bilimleri Dergisi (CFD), Cilt:36, No: 6 Özel Sayı (2015) ISSN: 1300-1949 Cumhuriyet University Faculty of Science Science Journal (CSJ), Vol. 36, No: 6 Special

More information

A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN)

A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN) A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN) Adeyinka A. Adewale, Samuel N. John, and Charles Ndujiuba 1 Department of Electrical and Information Engineering, Covenant

More information

A Fast Path Recovery Mechanism for MPLS Networks

A Fast Path Recovery Mechanism for MPLS Networks A Fast Path Recovery Mechanism for MPLS Networks Jenhui Chen, Chung-Ching Chiou, and Shih-Lin Wu Department of Computer Science and Information Engineering Chang Gung University, Taoyuan, Taiwan, R.O.C.

More information

How To Monitor And Test An Ethernet Network On A Computer Or Network Card

How To Monitor And Test An Ethernet Network On A Computer Or Network Card 3. MONITORING AND TESTING THE ETHERNET NETWORK 3.1 Introduction The following parameters are covered by the Ethernet performance metrics: Latency (delay) the amount of time required for a frame to travel

More information

Traffic Monitoring in a Switched Environment

Traffic Monitoring in a Switched Environment Traffic Monitoring in a Switched Environment InMon Corp. 1404 Irving St., San Francisco, CA 94122 www.inmon.com 1. SUMMARY This document provides a brief overview of some of the issues involved in monitoring

More information

Research on Video Traffic Control Technology Based on SDN. Ziyan Lin

Research on Video Traffic Control Technology Based on SDN. Ziyan Lin Joint International Mechanical, Electronic and Information Technology Conference (JIMET 2015) Research on Video Traffic Control Technology Based on SDN Ziyan Lin Communication University of China, Beijing

More information

Architecture of distributed network processors: specifics of application in information security systems

Architecture of distributed network processors: specifics of application in information security systems Architecture of distributed network processors: specifics of application in information security systems V.Zaborovsky, Politechnical University, Sait-Petersburg, Russia vlad@neva.ru 1. Introduction Modern

More information

Provisioning algorithm for minimum throughput assurance service in VPNs using nonlinear programming

Provisioning algorithm for minimum throughput assurance service in VPNs using nonlinear programming Provisioning algorithm for minimum throughput assurance service in VPNs using nonlinear programming Masayoshi Shimamura (masayo-s@isnaistjp) Guraduate School of Information Science, Nara Institute of Science

More information

2013 IEEE 14th International Conference on High Performance Switching and Routing

2013 IEEE 14th International Conference on High Performance Switching and Routing 03 IEEE 4th International Conference on High Performance Switching and Routing Cost and Delay Tradeoff in Three-Stage Switch Architecture for Data Center Networks Shu Fu, Bin Wu, Xiaohong Jiang *, Achille

More information

Load Balancing in Fault Tolerant Video Server

Load Balancing in Fault Tolerant Video Server Load Balancing in Fault Tolerant Video Server # D. N. Sujatha*, Girish K*, Rashmi B*, Venugopal K. R*, L. M. Patnaik** *Department of Computer Science and Engineering University Visvesvaraya College of

More information

Design and Implementation of Load-Balanced Multipath Self-routing Switching System

Design and Implementation of Load-Balanced Multipath Self-routing Switching System Proceedings of the International MultiConference of Engineers and Computer Scientists 24 Vol II, IMECS 24, March 2-4, 24, Hong Kong Design and Implementation of Load-Balanced Multipath Self-routing Switching

More information