The GLIMPS Terabit Packet Switching Engine

Size: px
Start display at page:

Download "The GLIMPS Terabit Packet Switching Engine"

Transcription

1 February 2002 The GLIMPS Terabit Packet Switching Engine I. Elhanany, O. Beeri Terabit Packet Switching Challenges The ever-growing demand for additional bandwidth reflects on the increasing capacity requirements from backbone switches and routers. As these systems increase in port count (tens and even hundreds) and line rates (oc-48 and oc-192), the engineering effort pertaining to the design and management of their switch fabric cores becomes more challenging. The performance of a switch/router as a whole is greatly affected by that of its fabric. Therefore, legacy packet-switching architecture and algorithms, which were suitable for the Gigabit space, need to be replaced by more scalable, higher-performance technologies. One of the primary building blocks of any switch fabric is the scheduler, which is in charge of determining which input ports are to transmit to which output ports at any given time. TeraCross patent-pending GLIMPS technology offers a unique and innovative packetscheduling scheme, which offers the scalability and performance characteristics, expected by tomorrow s high-end packet-switched networks. Current Trends in Switch Fabric Architectures Switches and routers are logically comprised of line cards and a switch fabric. On each line card, which is associated with a port, resides a network processing unit(s) that interfaces to the physical data links, decodes incoming traffic, finds the proper route for it, and applies traffic shaping, filtering and other policy functions. Hence the network processor is generally in charge of data received and transmitted by the port. The switch fabric is where actual transmission of data across ports takes place. In general, a fabric includes a set of crosspoint switch, a scheduler module and buffers. The crosspoints are configurable interconnecting elements that dynamically establish links between input and output ports. A central role is played by the scheduler which acts as the brain of the switch fabric; it configures the crosspoint, allows buffered data to traverse and repeatedly reconfigures the crosspoint for successive transmissions. Two fabric architectures are deployed

2 today at the mainstream of high-end switching systems: distributed shared-memory and crossbar based. SHARED MEMORY SWITCHES Let N denote the number of ports in a system. Shared memory schemes are very simple: each input port can send packets to each output port via a shared memory space in accordance with a prescribed arbitration scheme. Contention is resolved by speeding up the read/write operations to and from the shared memory device by N. In affect, this is an output queueing scheme since there is a dedicated logical path between any two ports. The downside of these schemes is the need for large speed up (order of N). This disadvantage prevents the scheme from being implemented in Terabit capacities. Yet another technique applicable to shared memory techniques, is packet-slicing over several shared memory chips. However, since such a scheme implies higher chip count and, as such, higher power consumption it is found to be limited in scalability. Utilizing a group of small shared memory switches to form a larger switch, such as in the use of a Clos Network topology, typically results in very high port count and large latencies. Such schemes also require complex management techniques, such as simultanteous configuration of the multiple shared-memory switches. Moreover, since power consumption is a direct product of the number of switching elements, shared memory solutions becomes less attractive for highend systems. SELF-ROUTING CROSSBARS A self-routing crossbar operates without global decision-making. Rather, it is based on multiple stages, each of which performs a basic switching task. Typically, a packet that enters into it propagates from the ingress node to the egress node of the crossbar without any need to configure the crossbar device by any external means. In most cases a self-routing crossbar is based on a net of basic modular crosspoints, each with a few ingress nodes and a few egress nodes (usually 2 for ingress and 2 for egress). To allow full connectivity the crossbar should have at least N log 2 N basic crosspoints (N crosspoints in each stage and log 2 N stages). When a packet enters the switch it receives a tag that defines the egress port it is destined to. This tag may be the definition of the exact path through the net or only the destination, in case of a worm hole algorithm.

3 The self-routing crossbar has some evident advantages over other schemes, such as modularity, which makes it easily scalable, and the fact that it is self-sustaining in the sense that no scheduler or other central management unit is required. However, a self-routing device has some obvious disadvantages and problems, resulting mainly from the fact that real data traffic is usually not free of contentions. When part of the path of two packets that enter the switch at the same time overlaps ( hot spot phenomena) only one of the packets may be forwarded while the other has to be dropped or buffered. Without buffering, this results in throughput degredation, which grows more prominent as the number of stages increase. Adding a buffer in the entry of each crosspoint will allow higher throughput, but is not cost-efficient and carries higher latency. Having the crossbar fully non-blocking (with N 2 crosspoints) does not solve the problem, rather it shifts it to the egress in the worst case the egress has to handle N packets simultaneously, which is impractical for high-end systems. To eliminate contentions inside the crossbar some additional mechanism is usually added, namely some sorting mechanism before the crossbar. Such a solution is obviously not the native and simple crossbar that we discussed so far, rather it requires a central management and scheduling unit that is added to the self-routing crossbar. Using simple sorting mechanism will, again, result in poor throughput, poor latency and very limited QoS supports. MULTISTAGE INTERCONNECT NETWORK A multistage interconnect network switch (MINS) is a switch that is based on a net of smaller interconnected sub-switches. This infrastructure is highly modular and theoretically allows growth almost to infinity. Since each of the sub-switches is relatively small, it may incorporate relatively complex and sophisticated scheduling algorithms, and therefore may solve scalability limitations. Nevertheless, using multistage switching has some overheads and disadvantages as will be discussed below. Each of the switch elements has only a partial view of the system state, which is limited to its local environment. This prevents taking decisions that are truly consistant with the overall system resources. Moreover, such local-scheduling decisions may overlap with each other, or worse, even contradict a desired global optimization. The controllability of the QoS provisioning in such a case is complex and is not straightforward. The full path of packets through the switch is the result of multi-routing decision and buffering, where each stage has its own overhead of queuing latency. The aggregation of all overheads is higher than in the case of single stage switching.

4 Although multistage interconnect switching is modular, the implementation for a high number of ports is not trivial, keeping in mind that up to 80% of the connectivity is allocated (thus wasted ) for interconnections. The overhead of multistage is therefore non-negligible. Moreover, since the I/O cost in space and power is of the same order as the logic cost, such structures are not cost-effective. CROSSBAR SWITCHES The crossbar is a configurable interconnecting element that dynamically establishes links between input and output ports. Nearly all crossbar devices up to date are electronic, however there has been significant progress in the optical domain, to provide all-optical high density crossbar modules. These are based on dynamic WDM, MEMS and other optical technologies. The clear advantages of using optical crossbars are the low-power, scalability and space efficiency. Electronic crossbar technology today offers over 128 input ports to be linked to 128 output ports each running at 3.125Gbps in a single chip package. This potential may only be realized if the crossbar is managed correctly. In other words, a central role is played by the scheduler which acts as the brain of the switch fabric; it configures the crossbar, allows buffered data to traverse it and repeatedly reconfigures the crossbar for successive transmissions. If the scheduler performs well, the entire fabric performs well. The task at hand, however, is far from being simple. Traditionally, switch fabrics were designed for moderate traffic loads that were considerably lower than the pragmatic router capacity. Such a design has the advantage of offering low average packet delay even when employing simple scheduling algorithms. As Internet traffic grows, it is necessary to increase the router capacity in order to maintain the ratio between traffic load and router capacity required for acceptable performance. Since the growth of traffic volume is faster than that of router capacity, simple algorithms are found inefficient and infeasible for modern routers. Internet traffic is accompanied by an increased volume of data which is sensitive to packet latency and jitter. Incoming packets are typically classified to Quality of Service (QoS) classes as means of providing differentiated services. For each QoS class the router must not exceed the acclaimed latency and jitter requirements. As can be derived, the task of the scheduler is a complex one: for each transmission period determine what packets from which input port are to be transmitted to their designated output ports. As high port count and QoS become predominent characteristics, the scheduler must complete its task within the time frame of a

5 single cell or packet, which for oc-192 translates to approximately 50 nsec - an extremely short period of time to complete such a compound task efficiently. Traffic Models Key to Performance Evaluation Traffic traversing real-life switches and routers is far from being ideal in the statistical sense. In other words, as opposed to being well scattered and uniformly distributed along the output ports, traffic tends to be burty and non-uniformly distributed. This is due to the diverse nature of applications and generating the traffic and the non-homogenous network aggregation. Advanced models, such as Self-Similar fractal models, have been recently proposed to represent the traffic flowing through the Internet backbone. These attempts, originating from hursitic and analytical work, indicate that the traffic is complex in behavior and should be regarded as such. To this end, the scheduling algorithms should be evaluated with respect to the traffic arriving at the systems. Frequently, various switching schemes are compared under ideal traffic as opposed to more complex, real-life scenarios. LEGACY SCHEDULING ALGORITHMS Commonly deployed scheduling algorithms are based on an early discipline, which was developed by DEC Systems Research Center for a 16-port, 1 Gbps switch, called Parallel Iterative Matching (PIM). The essence of PIM, as well as its popular derivatives, is to use randomness as means of avoiding starvation and maximizing the matching process. Inputs and outputs, which have not been previously matched, contend during each time slot. An iteration is comprised of the following three basic steps: Step 1 (Request) All unmatched inputs send requests to every output for which they have packets to be sent. Step 2 (Grant) - Each output selects at random one of its requesting inputs. Step 3 (Accept) - Each input selects at random a single output among those that granted it. The PIM technique has two interesting properties. First, by eliminating the matched pairs in each iteration convergence to a maximal match is assured in O(log N) iterations. Second, it ensures that all requests are eventually granted. PIM does, however, introduce some significant

6 drawbacks the principal of which is large queueing latencies in the presence of traffic loads exceeding 60%. Moreover, its inability to provide prioritized QoS and a requirement for O(N 2 ) connectivity renders PIM implementation impractical for modern switching cores. The basic round-robin matching (RRM) algorithm was designed to overcome the disadvantages of PIM in terms of both fairness and complexity. Instead of arbitrating randomly in steps 2 and 3, selections are made based on a prescribed rotating priority discipline. Two pointers are updated after every grant and every accept, respectively. RRM offers a minor improvement with respect to PIM whereas under non-uniformly distributed traffic loads the overall performance remains poor mainly due to a pointer synchronization phenomenon. A popular descendant of PIM and RRM, which was developed by Nick McKeown of Standford University, is a widely implemented scheduling algorithm called islip. islip is an iterative algorithm consisting of the following 3 steps: Step 1 (Request) All unmatched inputs send requests to every output for which they have packets to be sent. Step 2 (Grant) - Each output selects a requesting input coinciding with a predefined priority sequence. A pointer is utilized in order to indicate the current location of the highest priority elements. The pointer is incremented (modulo N) to one beyond the granted input if the latter is accepted in step 3. Step 3 (Accept) In a similar fashion to step 3, each input selects one of the granting outputs of step 2 in accordance with a predefined priority order. Similar to step 2, a unique pointer is maintained to indicate the position of the highest priority element. This pointer is incremented (modulo N) to one location beyond the accepted output. As opposed to updating the output pointer after every grant, such update occurs only if the grant has been accepted by an input in step 3. The islip algorithm has been shown to exhibit a significant reduction in pointer synchronization and accordingly higher throughput with a lower average packet delay. The algorithm does, however, convey some basic shortcomings including degraded performance in the presence of non-uniform and bursty traffic flows, lack of inherent QoS support and limited scalability with respect to high (>32) port densities. Almost all switch fabrics today are scheduled using some variant of the islip algorithm. These solutions are limited to the sub-320gbps space and could not scale to Terabit systems, much less support QoS and other such advanced features. The major drawback of request-grantaccept based algorithms, such as PIM and islip, lies in the independent treatment of queueing pointers which tend to synchronize, leading to degradation in the overall performance.

7 GLIMPS TECHNOLOGY BREAKING THE SCALABILITY BARRIER It would be intuitive to assume that the ideal scheduler consist of a centralized decision making unit which receives requests from all ports, determines the optimal matches and issues them to the crossbar and line cards. The only problem is that a for port counts and rates characterizing a Terabit router, such a centralized approach is found impractical, mainly due to the amount of information which needs to be sent from the line cards and received at the centralized scheduler. For this reason, it has been assumed so far that the distributed, pointer-based schemes are the only pragmatic schedluling approach. TeraCross patent-pending GLIMPS technology shatters this notion by offering an innovative approach to implementing a centralized scheduling mechanism which scales to multi- Terabit capacities. This is done due to a unique queueing architecture together with a matching packet-scheduling algorithm which together offer the following unprecedented features: Scalability GLIMPS architecture obeys an order NlogN complexity rule, which renders it practical to support hundreds of ports, hence multi-terabit capacities Low latency and jitter due to the centralized approach, the matching decisions are made with a global optimization perspective which translates to extremely low packet latencies and guarenteed bounded jitter Traffic invariance the GLIMPS algorithm would regard highly bursty and nonuniform traffic as it would ideal traffic, since the decision making process remains the same. As a result, performance metrics are almost unaffected by the statistical nature of the traffic thereby offering the inherent support of multi-media and other time-critical applications QoS provisioning the classes of service, as implemented in the ingress queueing, are projected to the GLIMPS algorithm in order to prioritize the various traffic flows. To this end, the support of differentiated service is an intergrated part of the decision making progress thereby offering QoS at the foundations of the fabric. Fairness centralized scheduling allows for assymetric ports to be regarded respectfully, thereby sharing the switching bandwidth allocation in a balanced and fair fashion No Starvation by reflecting an overall view of the queue status, the scheduler can verify that no queues remain unserved for prolonged periods of time. These features render TeraCross GLIMPS scheduling technology well positioned to lead the emergence of next-generation, multi-terabit, high-performance switch fabrics.

Efficient Queuing Architecture for a Buffered Crossbar Switch

Efficient Queuing Architecture for a Buffered Crossbar Switch Proceedings of the 11th WSEAS International Conference on COMMUNICATIONS, Agios Nikolaos, Crete Island, Greece, July 26-28, 2007 95 Efficient Queuing Architecture for a Buffered Crossbar Switch MICHAEL

More information

A Pipelined Memory Management Algorithm for Distributed Shared Memory Switches

A Pipelined Memory Management Algorithm for Distributed Shared Memory Switches A Pipelined Memory Management Algorithm for Distributed Shared Memory Switches Xike Li, Student Member, IEEE, Itamar Elhanany, Senior Member, IEEE* Abstract The distributed shared memory (DSM) packet switching

More information

Scalable Schedulers for High-Performance Switches

Scalable Schedulers for High-Performance Switches Scalable Schedulers for High-Performance Switches Chuanjun Li and S Q Zheng Mei Yang Department of Computer Science Department of Computer Science University of Texas at Dallas Columbus State University

More information

Dynamic Scheduling Algorithm for input-queued crossbar switches

Dynamic Scheduling Algorithm for input-queued crossbar switches Dynamic Scheduling Algorithm for input-queued crossbar switches Mihir V. Shah, Mehul C. Patel, Dinesh J. Sharma, Ajay I. Trivedi Abstract Crossbars are main components of communication switches used to

More information

DESIGN OF EFFICIENT ROUTING ALGORITHM FOR CONGESTION CONTROL IN NOC

DESIGN OF EFFICIENT ROUTING ALGORITHM FOR CONGESTION CONTROL IN NOC DESIGN OF EFFICIENT ROUTING ALGORITHM FOR CONGESTION CONTROL IN NOC 1 Pawar Ruchira Pradeep M. E, E&TC Signal Processing, Dr. D Y Patil School of engineering, Ambi, Pune Email: 1 ruchira4391@gmail.com

More information

Optical Packet Switching

Optical Packet Switching Optical Packet Switching DEISNet Gruppo Reti di Telecomunicazioni http://deisnet.deis.unibo.it WDM Optical Network Legacy Networks Edge Systems WDM Links λ 1 λ 2 λ 3 λ 4 Core Nodes 2 1 Wavelength Routing

More information

Generic Architecture. EECS 122: Introduction to Computer Networks Switch and Router Architectures. Shared Memory (1 st Generation) Today s Lecture

Generic Architecture. EECS 122: Introduction to Computer Networks Switch and Router Architectures. Shared Memory (1 st Generation) Today s Lecture Generic Architecture EECS : Introduction to Computer Networks Switch and Router Architectures Computer Science Division Department of Electrical Engineering and Computer Sciences University of California,

More information

IV. PACKET SWITCH ARCHITECTURES

IV. PACKET SWITCH ARCHITECTURES IV. PACKET SWITCH ARCHITECTURES (a) General Concept - as packet arrives at switch, destination (and possibly source) field in packet header is used as index into routing tables specifying next switch in

More information

Queuing. Congestion Control and Resource Allocation. Resource Allocation Evaluation Criteria. Resource allocation Drop disciplines Queuing disciplines

Queuing. Congestion Control and Resource Allocation. Resource Allocation Evaluation Criteria. Resource allocation Drop disciplines Queuing disciplines Resource allocation Drop disciplines Queuing disciplines Queuing 1 Congestion Control and Resource Allocation Handle congestion if and when it happens TCP Congestion Control Allocate resources to avoid

More information

EECS 122: Introduction to Computer Networks Switch and Router Architectures. Today s Lecture

EECS 122: Introduction to Computer Networks Switch and Router Architectures. Today s Lecture EECS : Introduction to Computer Networks Switch and Router Architectures Computer Science Division Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley,

More information

Unit 2 Packet Switching Networks - II

Unit 2 Packet Switching Networks - II Unit 2 Packet Switching Networks - II Dijkstra Algorithm: Finding shortest path Algorithm for finding shortest paths N: set of nodes for which shortest path already found Initialization: (Start with source

More information

Achieving Distributed Buffering in Multi-path Routing using Fair Allocation

Achieving Distributed Buffering in Multi-path Routing using Fair Allocation Achieving Distributed Buffering in Multi-path Routing using Fair Allocation Ali Al-Dhaher, Tricha Anjali Department of Electrical and Computer Engineering Illinois Institute of Technology Chicago, Illinois

More information

DiffServ Architecture: Impact of scheduling on QoS

DiffServ Architecture: Impact of scheduling on QoS DiffServ Architecture: Impact of scheduling on QoS Abstract: Scheduling is one of the most important components in providing a differentiated service at the routers. Due to the varying traffic characteristics

More information

Routers: Forwarding EECS 122: Lecture 13

Routers: Forwarding EECS 122: Lecture 13 Input Port Functions Routers: Forwarding EECS 22: Lecture 3 epartment of Electrical Engineering and Computer Sciences University of California Berkeley Physical layer: bit-level reception ata link layer:

More information

Chapter 1. Introduction

Chapter 1. Introduction Chapter 1 Introduction In a packet-switched network, packets are buffered when they cannot be processed or transmitted at the rate they arrive. There are three main reasons that a router, with generic

More information

Optical Interconnection Networks in Data Centers: Recent Trends and Future Challenges

Optical Interconnection Networks in Data Centers: Recent Trends and Future Challenges Optical Interconnection Networks in Data Centers: Recent Trends and Future Challenges Speaker: Lin Wang Research Advisor: Biswanath Mukherjee Kachris C, Kanonakis K, Tomkos I. Optical interconnection networks

More information

Stretch-Optimal Scheduling for On-Demand Data Broadcasts

Stretch-Optimal Scheduling for On-Demand Data Broadcasts Stretch-Optimal Scheduling for On-Demand Data roadcasts Yiqiong Wu and Guohong Cao Department of Computer Science & Engineering The Pennsylvania State University, University Park, PA 6 E-mail: fywu,gcaog@cse.psu.edu

More information

206 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 1, FEBRUARY The RGA arbitration can also start from the output side like in DRR [13] and

206 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 1, FEBRUARY The RGA arbitration can also start from the output side like in DRR [13] and 206 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 1, FEBRUARY 2008 Matching From the First Iteration: An Iterative Switching Algorithm for an Input Queued Switch Saad Mneimneh Abstract An iterative

More information

White Paper Enabling Quality of Service With Customizable Traffic Managers

White Paper Enabling Quality of Service With Customizable Traffic Managers White Paper Enabling Quality of Service With Customizable Traffic s Introduction Communications networks are changing dramatically as lines blur between traditional telecom, wireless, and cable networks.

More information

Hardware Evolution in Data Centers

Hardware Evolution in Data Centers Hardware Evolution in Data Centers 2004 2008 2011 2000 2013 2014 Trend towards customization Increase work done per dollar (CapEx + OpEx) Paolo Costa Rethinking the Network Stack for Rack-scale Computers

More information

TOC: Switching & Forwarding

TOC: Switching & Forwarding TOC: Switching & Forwarding Why? Switching Techniques Switch Characteristics Switch Examples Switch Architectures Summary Why? Direct vs. Switched Networks: Single link Switches Direct Network Limitations:

More information

TOC: Switching & Forwarding

TOC: Switching & Forwarding TOC: Switching & Forwarding Why? Switching Techniques Switch Characteristics Switch Examples Switch Architectures Summary TOC Switching Why? Direct vs. Switched Networks: n links Single link Direct Network

More information

Routers: Forwarding EECS 122: Lecture 13

Routers: Forwarding EECS 122: Lecture 13 Routers: Forwarding EECS 122: Lecture 13 epartment of Electrical Engineering and Computer Sciences University of California Berkeley Router Architecture Overview Two key router functions: run routing algorithms/protocol

More information

Crossbar - example. Crossbar. Crossbar. Combination: Time-space switching. Simple space-division switch Crosspoints can be turned on or off

Crossbar - example. Crossbar. Crossbar. Combination: Time-space switching. Simple space-division switch Crosspoints can be turned on or off Crossbar Crossbar - example Simple space-division switch Crosspoints can be turned on or off i n p u t s sessions: (,) (,) (,) (,) outputs Crossbar Advantages: simple to implement simple control flexible

More information

Optical networking technology

Optical networking technology 1 Optical networking technology Technological advances in semiconductor products have essentially been the primary driver for the growth of networking that led to improvements and simplification in the

More information

Satellite-Based Cellular Backhaul in the Era of LTE

Satellite-Based Cellular Backhaul in the Era of LTE Satellite-Based Cellular Backhaul in the Era of LTE Introduction 3 Essential Technologies for 3G/LTE Backhauling over Satellite 6 Gilat s Solution SkyEdge II-c Capricorn 7 Why Ultra-fast TDMA is the Only

More information

CSCD 433/533 Advanced Networks Spring Lecture 22 Quality of Service

CSCD 433/533 Advanced Networks Spring Lecture 22 Quality of Service CSCD 433/533 Advanced Networks Spring 2016 Lecture 22 Quality of Service 1 Topics Quality of Service (QOS) Defined Properties Integrated Service Differentiated Service 2 Introduction Problem Overview Have

More information

THE CAPACITY demand of the early generations of

THE CAPACITY demand of the early generations of IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 55, NO. 3, MARCH 2007 605 Resequencing Worst-Case Analysis for Parallel Buffered Packet Switches Ilias Iliadis, Senior Member, IEEE, and Wolfgang E. Denzel Abstract

More information

On Scheduling Unicast and Multicast Traffic in High Speed Routers

On Scheduling Unicast and Multicast Traffic in High Speed Routers On Scheduling Unicast and Multicast Traffic in High Speed Routers Kwan-Wu Chin School of Electrical, Computer and Telecommunications Engineering University of Wollongong kwanwu@uow.edu.au Abstract Researchers

More information

1 Architectures of Internet Switches and Routers

1 Architectures of Internet Switches and Routers 1 Architectures of Internet Switches and Routers Xin Li, Lotfi Mhamdi, Jing Liu, Konghong Pun, and Mounir Hamdi The Hong-Kong University of Science & Technology. {lixin,lotfi,liujing, konghong,hamdi}@cs.ust.hk

More information

Configuration of Offset Time in Optical Burst Switching Networks for Delay Sensitive Traffic

Configuration of Offset Time in Optical Burst Switching Networks for Delay Sensitive Traffic Configuration of Offset Time in Optical Burst Switching Networks for Delay Sensitive Traffic Anupam Soni and Yatindra Nath Singh anusoni@iitk.ac.in,ynsingh@iitk.ac.in. Abstract In Optical Burst Switching

More information

Long Round-Trip Time Support with Shared-Memory Crosspoint Buffered Packet Switch

Long Round-Trip Time Support with Shared-Memory Crosspoint Buffered Packet Switch Long Round-Trip Time Support with Shared-Memory Crosspoint Buffered Packet Switch Ziqian Dong and Roberto Rojas-Cessa Department of Electrical and Computer Engineering New Jersey Institute of Technology

More information

Latency on a Switched Ethernet Network

Latency on a Switched Ethernet Network FAQ 07/2014 Latency on a Switched Ethernet Network RUGGEDCOM Ethernet Switches & Routers http://support.automation.siemens.com/ww/view/en/94772587 This entry is from the Siemens Industry Online Support.

More information

Episode 5. Scheduling and Traffic Management

Episode 5. Scheduling and Traffic Management Episode 5. Scheduling and Traffic Management Part 3 Baochun Li Department of Electrical and Computer Engineering University of Toronto Outline What is scheduling? Why do we need it? Requirements of a scheduling

More information

H3C S9500 QoS Technology White Paper

H3C S9500 QoS Technology White Paper H3C Key words: QoS, quality of service Abstract: The Ethernet technology is widely applied currently. At present, Ethernet is the leading technology in various independent local area networks (LANs), and

More information

K-Selector-Based Dispatching Algorithm for Clos-Network Switches

K-Selector-Based Dispatching Algorithm for Clos-Network Switches K-Selector-Based Dispatching Algorithm for Clos-Network Switches Mei Yang, Mayauna McCullough, Yingtao Jiang, and Jun Zheng Department of Electrical and Computer Engineering, University of Nevada Las Vegas,

More information

Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks

Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks X. Yuan, R. Melhem and R. Gupta Department of Computer Science University of Pittsburgh Pittsburgh, PA 156 fxyuan,

More information

Quality of Service (QoS)

Quality of Service (QoS) Quality of Service (QoS) The Internet was originally designed for best-effort service without guarantee of predictable performance. Best-effort service is often sufficient for a traffic that is not sensitive

More information

Pi-PIFO: A Scalable Pipelined PIFO Memory Management Architecture

Pi-PIFO: A Scalable Pipelined PIFO Memory Management Architecture Pi-PIFO: A Scalable Pipelined PIFO Memory Management Architecture Steven Young, Student Member, IEEE, Itamar Arel, Senior Member, IEEE, Ortal Arazi, Member, IEEE Networking Research Group Electrical Engineering

More information

LS Example 5 3 C 5 A 1 D

LS Example 5 3 C 5 A 1 D Lecture 10 LS Example 5 2 B 3 C 5 1 A 1 D 2 3 1 1 E 2 F G Itrn M B Path C Path D Path E Path F Path G Path 1 {A} 2 A-B 5 A-C 1 A-D Inf. Inf. 1 A-G 2 {A,D} 2 A-B 4 A-D-C 1 A-D 2 A-D-E Inf. 1 A-G 3 {A,D,G}

More information

1 Introduction

1 Introduction Published in IET Communications Received on 17th September 2009 Revised on 3rd February 2010 ISSN 1751-8628 Multicast and quality of service provisioning in parallel shared memory switches B. Matthews

More information

Quality of Service in the Internet

Quality of Service in the Internet Quality of Service in the Internet Problem today: IP is packet switched, therefore no guarantees on a transmission is given (throughput, transmission delay, ): the Internet transmits data Best Effort But:

More information

Basic Switch Organization

Basic Switch Organization NOC Routing 1 Basic Switch Organization 2 Basic Switch Organization Link Controller Used for coordinating the flow of messages across the physical link of two adjacent switches 3 Basic Switch Organization

More information

Tag Switching. Background. Tag-Switching Architecture. Forwarding Component CHAPTER

Tag Switching. Background. Tag-Switching Architecture. Forwarding Component CHAPTER CHAPTER 23 Tag Switching Background Rapid changes in the type (and quantity) of traffic handled by the Internet and the explosion in the number of Internet users is putting an unprecedented strain on the

More information

Quality of Service in the Internet

Quality of Service in the Internet Quality of Service in the Internet Problem today: IP is packet switched, therefore no guarantees on a transmission is given (throughput, transmission delay, ): the Internet transmits data Best Effort But:

More information

CSE 123A Computer Networks

CSE 123A Computer Networks CSE 123A Computer Networks Winter 2005 Lecture 8: IP Router Design Many portions courtesy Nick McKeown Overview Router basics Interconnection architecture Input Queuing Output Queuing Virtual output Queuing

More information

Resource Sharing for QoS in Agile All Photonic Networks

Resource Sharing for QoS in Agile All Photonic Networks Resource Sharing for QoS in Agile All Photonic Networks Anton Vinokurov, Xiao Liu, Lorne G Mason Department of Electrical and Computer Engineering, McGill University, Montreal, Canada, H3A 2A7 E-mail:

More information

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists 3,900 6,000 20M Open access books available International authors and editors Downloads Our authors

More information

Introduction. Introduction. Router Architectures. Introduction. Recent advances in routing architecture including

Introduction. Introduction. Router Architectures. Introduction. Recent advances in routing architecture including Router Architectures By the end of this lecture, you should be able to. Explain the different generations of router architectures Describe the route lookup process Explain the operation of PATRICIA algorithm

More information

Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch

Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch Neda Beheshti, Nick Mckeown Stanford University Abstract In all internet routers buffers are needed to hold packets during times of congestion.

More information

CS 552 Computer Networks

CS 552 Computer Networks CS 55 Computer Networks IP forwarding Fall 00 Rich Martin (Slides from D. Culler and N. McKeown) Position Paper Goals: Practice writing to convince others Research an interesting topic related to networking.

More information

MAC Sublayer(1) Principal service of the Medium Access Control Sublayer: Allocating a single broadcast channel (mostly a LAN) among competing users

MAC Sublayer(1) Principal service of the Medium Access Control Sublayer: Allocating a single broadcast channel (mostly a LAN) among competing users MAC Sublayer(1) Principal service of the Medium Access Control Sublayer: Allocating a single broadcast channel (mostly a LAN) among competing users Static Channel Allocation: Frequency Division Multiplexing

More information

PARALLEL ALGORITHMS FOR IP SWITCHERS/ROUTERS

PARALLEL ALGORITHMS FOR IP SWITCHERS/ROUTERS THE UNIVERSITY OF NAIROBI DEPARTMENT OF ELECTRICAL AND INFORMATION ENGINEERING FINAL YEAR PROJECT. PROJECT NO. 60 PARALLEL ALGORITHMS FOR IP SWITCHERS/ROUTERS OMARI JAPHETH N. F17/2157/2004 SUPERVISOR:

More information

CHAPTER 6 FPGA IMPLEMENTATION OF ARBITERS ALGORITHM FOR NETWORK-ON-CHIP

CHAPTER 6 FPGA IMPLEMENTATION OF ARBITERS ALGORITHM FOR NETWORK-ON-CHIP 133 CHAPTER 6 FPGA IMPLEMENTATION OF ARBITERS ALGORITHM FOR NETWORK-ON-CHIP 6.1 INTRODUCTION As the era of a billion transistors on a one chip approaches, a lot of Processing Elements (PEs) could be located

More information

A New Integrated Unicast/Multicast Scheduler for Input-Queued Switches

A New Integrated Unicast/Multicast Scheduler for Input-Queued Switches Proc. 8th Australasian Symposium on Parallel and Distributed Computing (AusPDC 20), Brisbane, Australia A New Integrated Unicast/Multicast Scheduler for Input-Queued Switches Kwan-Wu Chin School of Electrical,

More information

Exam HP0-Y43 Implementing HP Network Infrastructure Solutions Version: 10.0 [ Total Questions: 62 ]

Exam HP0-Y43 Implementing HP Network Infrastructure Solutions Version: 10.0 [ Total Questions: 62 ] s@lm@n HP Exam HP0-Y43 Implementing HP Network Infrastructure Solutions Version: 10.0 [ Total Questions: 62 ] Question No : 1 A customer requires an HP FlexCampus solution with a core that scales to 40/100G.

More information

Latency on a Switched Ethernet Network

Latency on a Switched Ethernet Network Page 1 of 6 1 Introduction This document serves to explain the sources of latency on a switched Ethernet network and describe how to calculate cumulative latency as well as provide some real world examples.

More information

Sections Describing Standard Software Features

Sections Describing Standard Software Features 30 CHAPTER This chapter describes how to configure quality of service (QoS) by using automatic-qos (auto-qos) commands or by using standard QoS commands. With QoS, you can give preferential treatment to

More information

Advanced Computer Networks

Advanced Computer Networks Advanced Computer Networks QoS in IP networks Prof. Andrzej Duda duda@imag.fr Contents QoS principles Traffic shaping leaky bucket token bucket Scheduling FIFO Fair queueing RED IntServ DiffServ http://duda.imag.fr

More information

Technical Notes. QoS Features on the Business Ethernet Switch 50 (BES50)

Technical Notes. QoS Features on the Business Ethernet Switch 50 (BES50) Technical Notes QoS Features on the Business Ethernet Switch 50 (BES50) Version: NN70000-004 issue 1.00 Date: February 3 rd, 2009 Status: Released Copyright 2009 Nortel Networks. All rights reserved. The

More information

Shared-Memory Combined Input-Crosspoint Buffered Packet Switch for Differentiated Services

Shared-Memory Combined Input-Crosspoint Buffered Packet Switch for Differentiated Services Shared-Memory Combined -Crosspoint Buffered Packet Switch for Differentiated Services Ziqian Dong and Roberto Rojas-Cessa Department of Electrical and Computer Engineering New Jersey Institute of Technology

More information

TDDD82 Secure Mobile Systems Lecture 6: Quality of Service

TDDD82 Secure Mobile Systems Lecture 6: Quality of Service TDDD82 Secure Mobile Systems Lecture 6: Quality of Service Mikael Asplund Real-time Systems Laboratory Department of Computer and Information Science Linköping University Based on slides by Simin Nadjm-Tehrani

More information

Introduction to ATM Technology

Introduction to ATM Technology Introduction to ATM Technology ATM Switch Design Switching network (N x N) Switching network (N x N) SP CP SP CP Presentation Outline Generic Switch Architecture Specific examples Shared Buffer Switch

More information

Routing, Routers, Switching Fabrics

Routing, Routers, Switching Fabrics Routing, Routers, Switching Fabrics Outline Link state routing Link weights Router Design / Switching Fabrics CS 640 1 Link State Routing Summary One of the oldest algorithm for routing Finds SP by developing

More information

The Network Layer and Routers

The Network Layer and Routers The Network Layer and Routers Daniel Zappala CS 460 Computer Networking Brigham Young University 2/18 Network Layer deliver packets from sending host to receiving host must be on every host, router in

More information

Quality of Service in the Internet. QoS Parameters. Keeping the QoS. Leaky Bucket Algorithm

Quality of Service in the Internet. QoS Parameters. Keeping the QoS. Leaky Bucket Algorithm Quality of Service in the Internet Problem today: IP is packet switched, therefore no guarantees on a transmission is given (throughput, transmission delay, ): the Internet transmits data Best Effort But:

More information

The Arbitration Problem

The Arbitration Problem HighPerform Switchingand TelecomCenterWorkshop:Sep outing ance t4, 97. EE84Y: Packet Switch Architectures Part II Load-balanced Switches ick McKeown Professor of Electrical Engineering and Computer Science,

More information

FM4000. A Scalable, Low-latency 10 GigE Switch for High-performance Data Centers

FM4000. A Scalable, Low-latency 10 GigE Switch for High-performance Data Centers A Scalable, Low-latency 10 GigE Switch for High-performance Data Centers Uri Cummings Rebecca Collins Virat Agarwal Dan Daly Fabrizio Petrini Michael Perrone Davide Pasetto Hot Interconnects 17 (Aug 2009)

More information

Master Course Computer Networks IN2097

Master Course Computer Networks IN2097 Chair for Network Architectures and Services Prof. Carle Department for Computer Science TU München Chair for Network Architectures and Services Prof. Carle Department for Computer Science TU München Master

More information

Mark Sandstrom ThroughPuter, Inc.

Mark Sandstrom ThroughPuter, Inc. Hardware Implemented Scheduler, Placer, Inter-Task Communications and IO System Functions for Many Processors Dynamically Shared among Multiple Applications Mark Sandstrom ThroughPuter, Inc mark@throughputercom

More information

Resource allocation in networks. Resource Allocation in Networks. Resource allocation

Resource allocation in networks. Resource Allocation in Networks. Resource allocation Resource allocation in networks Resource Allocation in Networks Very much like a resource allocation problem in operating systems How is it different? Resources and jobs are different Resources are buffers

More information

ECE 551 System on Chip Design

ECE 551 System on Chip Design ECE 551 System on Chip Design Introducing Bus Communications Garrett S. Rose Fall 2018 Emerging Applications Requirements Data Flow vs. Processing µp µp Mem Bus DRAMC Core 2 Core N Main Bus µp Core 1 SoCs

More information

Network Superhighway CSCD 330. Network Programming Winter Lecture 13 Network Layer. Reading: Chapter 4

Network Superhighway CSCD 330. Network Programming Winter Lecture 13 Network Layer. Reading: Chapter 4 CSCD 330 Network Superhighway Network Programming Winter 2015 Lecture 13 Network Layer Reading: Chapter 4 Some slides provided courtesy of J.F Kurose and K.W. Ross, All Rights Reserved, copyright 1996-2007

More information

CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS

CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS 28 CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS Introduction Measurement-based scheme, that constantly monitors the network, will incorporate the current network state in the

More information

Shared-Memory Combined Input-Crosspoint Buffered Packet Switch for Differentiated Services

Shared-Memory Combined Input-Crosspoint Buffered Packet Switch for Differentiated Services Shared-Memory Combined -Crosspoint Buffered Packet Switch for Differentiated Services Ziqian Dong and Roberto Rojas-Cessa Department of Electrical and Computer Engineering New Jersey Institute of Technology

More information

Efficient Multicast Support in Buffered Crossbars using Networks on Chip

Efficient Multicast Support in Buffered Crossbars using Networks on Chip Efficient Multicast Support in Buffered Crossbars using etworks on Chip Iria Varela Senin Lotfi Mhamdi Kees Goossens, Computer Engineering, Delft University of Technology, Delft, The etherlands XP Semiconductors,

More information

Worst-case Ethernet Network Latency for Shaped Sources

Worst-case Ethernet Network Latency for Shaped Sources Worst-case Ethernet Network Latency for Shaped Sources Max Azarov, SMSC 7th October 2005 Contents For 802.3 ResE study group 1 Worst-case latency theorem 1 1.1 Assumptions.............................

More information

Switching Using Parallel Input Output Queued Switches With No Speedup

Switching Using Parallel Input Output Queued Switches With No Speedup IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 10, NO. 5, OCTOBER 2002 653 Switching Using Parallel Input Output Queued Switches With No Speedup Saad Mneimneh, Vishal Sharma, Senior Member, IEEE, and Kai-Yeung

More information

The Design and Performance Analysis of QoS-Aware Edge-Router for High-Speed IP Optical Networks

The Design and Performance Analysis of QoS-Aware Edge-Router for High-Speed IP Optical Networks The Design and Performance Analysis of QoS-Aware Edge-Router for High-Speed IP Optical Networks E. Kozlovski, M. Düser, R. I. Killey, and P. Bayvel Department of and Electrical Engineering, University

More information

Quest for High-Performance Bufferless NoCs with Single-Cycle Express Paths and Self-Learning Throttling

Quest for High-Performance Bufferless NoCs with Single-Cycle Express Paths and Self-Learning Throttling Quest for High-Performance Bufferless NoCs with Single-Cycle Express Paths and Self-Learning Throttling Bhavya K. Daya, Li-Shiuan Peh, Anantha P. Chandrakasan Dept. of Electrical Engineering and Computer

More information

Cisco ASR 1000 Series Aggregation Services Routers: QoS Architecture and Solutions

Cisco ASR 1000 Series Aggregation Services Routers: QoS Architecture and Solutions Cisco ASR 1000 Series Aggregation Services Routers: QoS Architecture and Solutions Introduction Much more bandwidth is available now than during the times of 300-bps modems, but the same business principles

More information

Introduction. Router Architectures. Introduction. Introduction. Recent advances in routing architecture including

Introduction. Router Architectures. Introduction. Introduction. Recent advances in routing architecture including Introduction Router Architectures Recent advances in routing architecture including specialized hardware switching fabrics efficient and faster lookup algorithms have created routers that are capable of

More information

Managed IP Services from Dial Access to Gigabit Routers

Managed IP Services from Dial Access to Gigabit Routers Managed IP Services from Dial Access to Gigabit Routers Technical barriers and Future trends for IP Differentiated Services Grenville Armitage, PhD Member of Technical Staff High Speed Networks Research,

More information

Module 17: "Interconnection Networks" Lecture 37: "Introduction to Routers" Interconnection Networks. Fundamentals. Latency and bandwidth

Module 17: Interconnection Networks Lecture 37: Introduction to Routers Interconnection Networks. Fundamentals. Latency and bandwidth Interconnection Networks Fundamentals Latency and bandwidth Router architecture Coherence protocol and routing [From Chapter 10 of Culler, Singh, Gupta] file:///e /parallel_com_arch/lecture37/37_1.htm[6/13/2012

More information

Introduction to Cisco ASR 9000 Series Network Virtualization Technology

Introduction to Cisco ASR 9000 Series Network Virtualization Technology White Paper Introduction to Cisco ASR 9000 Series Network Virtualization Technology What You Will Learn Service providers worldwide face high customer expectations along with growing demand for network

More information

Chapter 7 CONCLUSION

Chapter 7 CONCLUSION 97 Chapter 7 CONCLUSION 7.1. Introduction A Mobile Ad-hoc Network (MANET) could be considered as network of mobile nodes which communicate with each other without any fixed infrastructure. The nodes in

More information

Design and Evaluation of a Parallel-Polled Virtual Output Queued Switch *

Design and Evaluation of a Parallel-Polled Virtual Output Queued Switch * Design and Evaluation of a Parallel-Polled Virtual Output Queued Switch * K. J. Christensen Department of Computer Science and Engineering University of South Florida Tampa, FL 3360 Abstract - Input-buffered

More information

Priority Traffic CSCD 433/533. Advanced Networks Spring Lecture 21 Congestion Control and Queuing Strategies

Priority Traffic CSCD 433/533. Advanced Networks Spring Lecture 21 Congestion Control and Queuing Strategies CSCD 433/533 Priority Traffic Advanced Networks Spring 2016 Lecture 21 Congestion Control and Queuing Strategies 1 Topics Congestion Control and Resource Allocation Flows Types of Mechanisms Evaluation

More information

Designing of Efficient islip Arbiter using islip Scheduling Algorithm for NoC

Designing of Efficient islip Arbiter using islip Scheduling Algorithm for NoC International Journal of Scientific and Research Publications, Volume 3, Issue 12, December 2013 1 Designing of Efficient islip Arbiter using islip Scheduling Algorithm for NoC Deepali Mahobiya Department

More information

Basics (cont.) Characteristics of data communication technologies OSI-Model

Basics (cont.) Characteristics of data communication technologies OSI-Model 48 Basics (cont.) Characteristics of data communication technologies OSI-Model Topologies Packet switching / Circuit switching Medium Access Control (MAC) mechanisms Coding Quality of Service (QoS) 49

More information

Master Course Computer Networks IN2097

Master Course Computer Networks IN2097 Chair for Network Architectures and Services Prof. Carle Department for Computer Science TU München Master Course Computer Networks IN2097 Prof. Dr.-Ing. Georg Carle Christian Grothoff, Ph.D. Chair for

More information

Toward a Reliable Data Transport Architecture for Optical Burst-Switched Networks

Toward a Reliable Data Transport Architecture for Optical Burst-Switched Networks Toward a Reliable Data Transport Architecture for Optical Burst-Switched Networks Dr. Vinod Vokkarane Assistant Professor, Computer and Information Science Co-Director, Advanced Computer Networks Lab University

More information

Lecture 9. Quality of Service in ad hoc wireless networks

Lecture 9. Quality of Service in ad hoc wireless networks Lecture 9 Quality of Service in ad hoc wireless networks Yevgeni Koucheryavy Department of Communications Engineering Tampere University of Technology yk@cs.tut.fi Lectured by Jakub Jakubiak QoS statement

More information

Lossless 10 Gigabit Ethernet: The Unifying Infrastructure for SAN and LAN Consolidation

Lossless 10 Gigabit Ethernet: The Unifying Infrastructure for SAN and LAN Consolidation . White Paper Lossless 10 Gigabit Ethernet: The Unifying Infrastructure for SAN and LAN Consolidation Introduction As organizations increasingly rely on IT to help enable, and even change, their business

More information

Integrated Scheduling and Buffer Management Scheme for Input Queued Switches under Extreme Traffic Conditions

Integrated Scheduling and Buffer Management Scheme for Input Queued Switches under Extreme Traffic Conditions Integrated Scheduling and Buffer Management Scheme for Input Queued Switches under Extreme Traffic Conditions Anuj Kumar, Rabi N. Mahapatra Texas A&M University, College Station, U.S.A Email: {anujk, rabi}@cs.tamu.edu

More information

FIRM: A Class of Distributed Scheduling Algorithms for High-speed ATM Switches with Multiple Input Queues

FIRM: A Class of Distributed Scheduling Algorithms for High-speed ATM Switches with Multiple Input Queues FIRM: A Class of Distributed Scheduling Algorithms for High-speed ATM Switches with Multiple Input Queues D.N. Serpanos and P.I. Antoniadis Department of Computer Science University of Crete Knossos Avenue

More information

A MULTIPROCESSOR SYSTEM. Mariam A. Salih

A MULTIPROCESSOR SYSTEM. Mariam A. Salih A MULTIPROCESSOR SYSTEM Mariam A. Salih Multiprocessors classification. interconnection networks (INs) Mode of Operation Control Strategy switching techniques Topology BUS-BASED DYNAMIC INTERCONNECTION

More information

Providing Flow Based Performance Guarantees for Buffered Crossbar Switches

Providing Flow Based Performance Guarantees for Buffered Crossbar Switches Providing Flow Based Performance Guarantees for Buffered Crossbar Switches Deng Pan Dept. of Electrical & Computer Engineering Florida International University Miami, Florida 33174, USA pand@fiu.edu Yuanyuan

More information

Modular Quality of Service Overview on Cisco IOS XR Software

Modular Quality of Service Overview on Cisco IOS XR Software Modular Quality of Service Overview on Cisco IOS XR Software Quality of Service (QoS) is the technique of prioritizing traffic flows and providing preferential forwarding for higher-priority packets. The

More information

Wireless Networks (CSC-7602) Lecture 8 (15 Oct. 2007)

Wireless Networks (CSC-7602) Lecture 8 (15 Oct. 2007) Wireless Networks (CSC-7602) Lecture 8 (15 Oct. 2007) Seung-Jong Park (Jay) http://www.csc.lsu.edu/~sjpark 1 Today Wireline Fair Schedulling Why? Ideal algorithm Practical algorithms Wireless Fair Scheduling

More information