Khiem Lam Jimmy Vuong Andrew Yang

Slides:



Advertisements
Similar presentations
Winter 2004 UCSC CMPE252B1 CMPE 257: Wireless and Mobile Networking SET 3f: Medium Access Control Protocols.
Advertisements

Wireless Resource Management through Packet Scheduling Outline for this lecture o identify the design challenges for QoS support over wireless mobile networks.
Maximum Battery Life Routing to Support Ubiquitous Mobile Computing in Wireless Ad Hoc Networks By C. K. Toh.
2005/12/06OPLAB, Dept. of IM, NTU1 Optimizing the ARQ Performance in Downlink Packet Data Systems With Scheduling Haitao Zheng, Member, IEEE Harish Viswanathan,
1 A Novel Topology-blind Fair Medium Access Control for Wireless LAN and Ad Hoc Networks Z. Y. Fang and B. Bensaou Computer Science Department Hong Kong.
Worst-case Fair Weighted Fair Queueing (WF²Q) by Jon C.R. Bennett & Hui Zhang Presented by Vitali Greenberg.
Generalized Processing Sharing (GPS) Is work conserving Is a fluid model Service Guarantee –GPS discipline can provide an end-to-end bounded- delay service.
Service Disciplines for Guaranteed Performance Service Hui Zhang, “Service Disciplines for Guaranteed Performance Service in Packet-Switching Networks,”
1 Token Bucket Based CAC and Packet Scheduling for IEEE Broadband Wireless Access Networks Chi-Hung Chiang
An Overview of Scheduling Algorithms in Wireless Multimedia Networks Hossam Fattah, Cyril Leung (The University of British Columbia) presented by Metin.
Distributed Fair Scheduling in a Wireless LAN Gautam Kulkarni EE206A (Spring 2001) Nitin Vaidya, Paramvir Bahl and Seema Gupta (appeared in Mobicom 2000.
1 Short-term Fairness for TCP Flows in b WLANs M. Bottigliengo, C. Casetti, C.-F. Chiasserini, M. Meo INFOCOM 2004.
Fair Real-time Traffic Scheduling over Wireless Local Area Networks Insik Shin Joint work with M. Adamou, S. Khanna, I. Lee, and S. Zhou Dept. of Computer.
21 Mar 2002 First Israelli-Swedish Workshop on Next Generation Networking1 / 28 Stochastic Analysis of Wireless-fair Scheduling Hwee Pink Tan and Raphael.
Company LOGO Provision of Multimedia Services in based Networks Colin Roby CMSC 681 Fall 2007.
CONGESTION CONTROL and RESOURCE ALLOCATION. Definition Resource Allocation : Process by which network elements try to meet the competing demands that.
Computer Networks Performance Metrics. Performance Metrics Outline Generic Performance Metrics Network performance Measures Components of Hop and End-to-End.
Fair Queueing. 2 First-Come-First Served (FIFO) Packets are transmitted in the order of their arrival Advantage: –Very simple to implement Disadvantage:
Admission Control and Scheduling for QoS Guarantees for Variable-Bit-Rate Applications on Wireless Channels I-Hong Hou P.R. Kumar University of Illinois,
Incentive-Oriented Downlink Scheduling for Wireless Networks with Real-Time and Non-Real-Time Flows I-Hong Hou, Jing Zhu, and Rath Vannithamby.
Packet Scheduling: SCFQ, STFQ, WF2Q Yongho Seok Contents Review: GPS, PGPS SCFQ( Self-clocked fair queuing ) STFQ( Start time fair queuing ) WF2Q( Worst-case.
Dynamic Bandwidth Allocation with Fair Scheduling For WCDMA Systems Liang Xu, Xumin Shen, and Jon W. Mark University of Waterloo published in IEEE Wireless.
Scheduling Determines which packet gets the resource. Enforces resource allocation to each flows. To be “Fair”, scheduling must: –Keep track of how many.
Content caching and scheduling in wireless networks with elastic and inelastic traffic Group-VI 09CS CS CS30020 Performance Modelling in Computer.
1 Fair Queuing Hamed Khanmirza Principles of Network University of Tehran.
1 A Cross-Layer Scheduling Algorithm With QoS Support in Wireless Networks Qingwen Liu, Student Member, IEEE, Xin Wang, Member, IEEE, and Georgios B. Giannakis,
Queue Scheduling Disciplines
Courtesy Piggybacking: Supporting Differentiated Services in Multihop Mobile Ad Hoc Networks Wei LiuXiang Chen Yuguang Fang WING Dept. of ECE University.
Providing QoS in IP Networks
Scheduling for QoS Management. Engineering Internet QoS2 Outline  What is Queue Management and Scheduling?  Goals of scheduling  Fairness (Conservation.
-1/16- Maximum Battery Life Routing to Support Ubiquitous Mobile Computing in Wireless Ad Hoc Networks C.-K. Toh, Georgia Institute of Technology IEEE.
04/02/08 1 Packet Scheduling IT610 Prof. A. Sahoo KReSIT.
Puzzle You have 2 glass marbles Building with 100 floors
Wireless MAC.
Instructor Materials Chapter 6: Quality of Service
Team: Aaron Sproul Patrick Hamilton
Balancing Uplink and Downlink Delay of VoIP Traffic in WLANs
Topics discussed in this section:
Presented by Tae-Seok Kim
Nathan Daniel Anil Koneri Vineeth Chander Yuhang Lin Jaime Johnson
Sriram Lakshmanan Zhenyun Zhuang
Wireless Fair Scheduling
Wireless Scheduling.
Wireless Scheduling.
HCF and EDCF Simulations
Wireless MAC.
Measuring Service in Multi-Class Networks
Stratified Round Robin: A Low Complexity Packet Scheduler with Bandwidth Fairness and Bounded Delay Sriram Ramabhadran Joseph Pasquale Presented by Sailesh.
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 6: Quality of Service Connecting Networks.
Lottery Scheduling Ish Baid.
Howard Huang, Sivarama Venkatesan, and Harish Viswanathan
Quality of Service For Traffic Aggregates
Dusit Niyato, Student Member, IEEE Ekram Hossain, Senior Member, IEEE
Variations of Weighted Fair Queueing
Provision of Multimedia Services in based Networks
Scheduling Algorithms in Broad-Band Wireless Networks
Lottery Meets Wireless
Fair Queueing.
Variations of Weighted Fair Queueing
Congestion Control Reasons:
Subject Name: Adhoc Networks Subject Code: 10CS841
Youngki Kim Mobile R&D Laboratory KT, Korea
Chapter-5 Traffic Engineering.
Introduction to Packet Scheduling
Modeling and Evaluating Variable Bit rate Video Steaming for ax
Fair Queueing.
A Simple QoS Packet Scheduler for Network Routers
Introduction to Packet Scheduling
Towards Predictable Datacenter Networks
Presentation transcript:

Khiem Lam Jimmy Vuong Andrew Yang A Unified Architecture for the Design and Evaluation of Wireless Fair Queueing Algorithms Khiem Lam Jimmy Vuong Andrew Yang ECE4605 – Fall 2007

Outline Present wireless fair service model Present unified wireless fair queuing architecture Evaluate and compare these 7 algorithms

Problems Location-dependent errors Bursty errors Channel contention Joint scheduling of uplink and downlink flows

Example Assume you have a packet cellular network with each base station performing the scheduling of the uplink and downlink transmissions, where communication can be only downlink or only uplink. Multiple channels are used and every mobile host can communicate with the base station.

Example Key characteristics of this wireless channel: Wireless channel capacity is dynamically varying Channel errors are location-dependent and bursty in nature There is contention in the channel among multiple mobile hosts Mobile hosts do not have global channel state Scheduling deals with both uplink and downlink flows Mobile hosts are often constrained in terms of processing power and battery power

Properties of Fluid Fair Queuing Fairness among backlogged flows Bounded delay channel access Guaranteed minimum throughput for backlogged flows In order to keep these properties, wireless channels require some properties: Basically, fluid fair queuing provides a separation between flows so each flows don’t affect each other. Ex minimum flow guarantees are unaffected by the behavior of other flows. Fluid fair queuing assumes that the channel is error free or at least not location dependent. However, when there is location dependent errors, then it can’t provide fair or minimum throughput bounds.

Fair Queuing in Wireless Channels Properties: Short-term fairness Long-term fairness Channel-conditioned delay bounds Short-term throughput bounds Long-term throughput bounds Delay sensitive and error sensitive data flows Optimization of schedulable region

Issues in Wireless Fair Queuing There are three major issues that need to be addressed: The failure of traditional fluid fair queuing in the presence of location-dependent channel error The compensation model for flows that preceive channel error: how transparent should wireless channel errors be to the user? The trade off between full separation and compensation, and its impact on fairness of channel access

Five Components of Wireless Fair Queuing Architecture Error-free service Lead and lag model Compensation model Slot queues and packet queues Channel monitoring and prediction Goal is to emulate fluid fair queuing when channel is error free and when there are location dependent errors, swap clean channels with location error channels and vice versa. Error-free service: Defines ideal fair service model assuming no channel errors Lead and lag model: Determines which flows are leading/logging their error free service and by how much Compensation model: compensates lagging flows that perceive an error-free channel at the expensive of leading flows, so it solves bursty and location dependent channel error Slot queues and packet queues: allows support of delay sensitive and error sensitive flows; also decouples connection-level packet management policies from link level packet scheduling policies Channel monitoring and prediction: provides reliable and accurate measurement and estimation of the channel state at any time instant for each backlogged flow

Error-free Service Model Reference for how much service a flow should ideally receive Packetized approximation of fluid fair queuing Lead and Lag Model Amount of service that can be compensated Leadering, lagging, and in sync flows may change dynamically over time Lag of a lagging flow = amount of additional service to which it is entitled in the future to compensate for lost service in the past Lead of a leading flow amount of additional service that the flow can give up to compensative for the additional service it received in the past. 2 ways to compute lag and lead (see end of 3.2)

Compensation Model Purpose is to reclaim the lost service due to channel error Leading flows give up their excess service gradually 4 Situations: No compensation Flow with maximum lag Leading and lagging flows swap slots Bandwidth reserved for compensation

7 Wireless Fair Queuing Algorithms Channel State Dependent Packet Scheduling (CSDPS) Idealized Wireless Fair Queueing (IWFQ) Wireless Packet Scheduling (WPS) Chanel-condition Independent Fair Queueing (CIF-Q) Enhanced Class Based Queueing with CSDPS (CBQ- CSDPS) Server Based Fairness Approach (SBFA) Wireless Fair Service Algorithm (WFS)

Channel State Dependent Packet Scheduling (CSDPS) Error-free Service: WRR Lead/Lag: Handles only error-free flows Compensation: No measure of lead/lag Complexity: O(n) Simple Disregards flows that perceives error, handles only error-free service O(n) – checks if each flow is backlogged and perceives a clean channel

Idealized Wireless Fair Queuing (IWPQ) Error-free Service: WFQ Lead/Lag: Calculate lead/lag by comparing service tag to error-free service tag Compensation: Lowest service tag gets priority to transmit Complexity: O(nlogn) Service tag is set to the finish tag of its head-of-line packet O(nlogn) – sorting service tags

Wireless Packet Scheduling (WPS) Error-free Service: WRR with WFQ Lead/Lag: Increment/decrement lag of flows that are swapped. Compensation: Intra-frame swapping and generating new frames base on “effective weight” flows Complexity: O(n) Intra-frame swapping – compensate flows that encounter channel error by locally trading slot allocations. Lead/lag accounting mechanism – provide additional services tot the lagging flows at the expense of the leading flows by

Channel-condition Independent Fair Queuing (CIF-Q) Error-free Service: Start Time Fair Queuing Lead/Lag: Calculate lead/lag by comparing service tag to error-free service tag Compensation: Leading flow relinquishes slots for lagging flows Complexity: O(nlogn) STFQ – similar to WFQ Relinquished Slots – distributed among lagging flows in proportion to the lagging flows’ rate weights. O(nlogn) – to sort service tags

Enhanced Class Based Queuing with CSDPS (CBQ-CSDPS) Error-free Service: Modified CBQ with CSDPS Lead/Lag: Based on actual number of bytes transmitted during each time window Compensation Model: Lagging flows are given explicit precedence. Complexity: O(n) Lead/Lag: compared to normalized weight. Leading excess compared to normal. Lagging less than normal. O(n) – same as CSDPS, to check if each flow is backlogged and perceives a clean channel

Server Based Fairness Approach (SBFA) Error-free Service: Generic framework for adapting different service disciplines Lead/Lag: Lag is measured by the number of slots in compensation flow (no measure for leading flows) Compensation Model: Compensation flow treated like any other flow Complexity: Dependent on error-free service Sensitive to statically reserved fraction

Wireless Fair Service (WFS) Error-free service: Enhanced WFQ to support delay- bandwidth decoupling Lead/Lag: Lead is the # of slots it gives up in the future whereas lag is the # of slots the flow is entitled to make up in the future. Compensation Model: Leading flow with lead, l, and lead bound of lmax relinquishes l/lmax slots allocated to it Complexity: O(nlogn) O(nlogn) – to sort service tags

Wireless Fair Queuing Algorithms Summary / - coarsely achieved *1 – only for error-free channels   Wireless Fair Queuing Algorithms Wireless Fair Service Model Properties CSDPS IWFQ WPS CIF-Q CBQ-CSDPS SBFA WFS Short-term Fairness / ✔ Long-term Fairness Channel-conditioned Delay Bounds Short-term Throughput Bounds Long-term Throughput Bounds ✔*1 Optimization of the Schedulable Region

Simulation Results Average result over 40 simulation runs CBR, Poisson, and MMPP sources Markov chain wireless channel Six Examples

Examples 1 & 2 Example 1: Error-free Service – All algorithms perform to their error-free models Example 2: Error-sensitive vs. Delay-sensitive flows

Example 3 – Service Degradation

Example 4, 5, & 6 Example 4: Channel Prediction Worse channel prediction, worse performance Example 5: Identical Behavior Moderately loaded network Moderate error patterns Large number of sources Example 6: Adaptive Sources Effect of latency adaptation on throughput in channel error

Related Work Adaptation for wireless LAN, WAN, PAN, etc. Adaptive sources (http, streaming) Forward Error Correction

Conclusion and Critique First work that provides evaluation of different wireless fair queuing algorithms WFS and CIF-Q achieve all properties of wireless fair service Paper needed more revising (typos, etc.) Paper did not give many details on algorithms