Draft-constantine-ippm-tcp-throughput-tm-02.txt 1 TCP Throughput Testing Methodology IETF 77 Anaheim Barry Constantine Reinhard.

Slides:



Advertisements
Similar presentations
Michele Pagano – A Survey on TCP Performance Evaluation and Modeling 1 Department of Information Engineering University of Pisa Network Telecomunication.
Advertisements

LOGO Transmission Control Protocol 12 (TCP) Data Flow.
CSIT560 Internet Infrastructure: Switches and Routers Active Queue Management Presented By: Gary Po, Henry Hui and Kenny Chong.
TELE202 Lecture 8 Congestion control 1 Lecturer Dr Z. Huang Overview ¥Last Lecture »X.25 »Source: chapter 10 ¥This Lecture »Congestion control »Source:
Top-Down Network Design Chapter Thirteen Optimizing Your Network Design Copyright 2010 Cisco Press & Priscilla Oppenheimer.
24-1 Chapter 24. Congestion Control and Quality of Service (part 1) 23.1 Data Traffic 23.2 Congestion 23.3 Congestion Control 23.4 Two Examples.
By Arjuna Sathiaseelan Tomasz Radzik Department of Computer Science King’s College London EPDN: Explicit Packet Drop Notification and its uses.
1 TCP Congestion Control. 2 TCP Segment Structure source port # dest port # 32 bits application data (variable length) sequence number acknowledgement.
Improving TCP Performance over Mobile Ad Hoc Networks by Exploiting Cross- Layer Information Awareness Xin Yu Department Of Computer Science New York University,
Congestion Control Created by M Bateman, A Ruddle & C Allison As part of the TCP View project.
School of Information Technologies TCP Congestion Control NETS3303/3603 Week 9.
Receiver-driven Layered Multicast S. McCanne, V. Jacobsen and M. Vetterli SIGCOMM 1996.
Transport Layer 3-1 Fast Retransmit r time-out period often relatively long: m long delay before resending lost packet r detect lost segments via duplicate.
TDC365 Spring 2001John Kristoff - DePaul University1 Internetworking Technologies Transmission Control Protocol (TCP)
Congestion Control Tanenbaum 5.3, /12/2015Congestion Control (A Loss Based Technique: TCP)2 What? Why? Congestion occurs when –there is no reservation.
Transport Layer3-1 Congestion Control. Transport Layer3-2 Principles of Congestion Control Congestion: r informally: “too many sources sending too much.
Networks: Congestion Control1 Congestion Control.
CSEE W4140 Networking Laboratory Lecture 7: TCP flow control and congestion control Jong Yul Kim
1 Chapter 3 Transport Layer. 2 Chapter 3 outline 3.1 Transport-layer services 3.2 Multiplexing and demultiplexing 3.3 Connectionless transport: UDP 3.4.
Data Communication and Networks
Semester Copyright USM EEE449 Computer Networks Congestion En. Mohd Nazri Mahmud MPhil (Cambridge, UK) BEng (Essex, UK) Room.
CMPE 257 Spring CMPE 257: Wireless and Mobile Networking Spring 2005 E2E Protocols (point-to-point)
Error Checking continued. Network Layers in Action Each layer in the OSI Model will add header information that pertains to that specific protocol. On.
Bandwidth Estimation: Metrics Mesurement Techniques and Tools By Ravi Prasad, Constantinos Dovrolis, Margaret Murray and Kc Claffy IEEE Network, Nov/Dec.
Ch. 28 Q and A IS 333 Spring Q1 Q: What is network latency? 1.Changes in delay and duration of the changes 2.time required to transfer data across.
Draft-constantine-ippm-tcp-throughput-tm-00.txt 1 TCP Throughput Testing Methodology IETF 76 Hiroshima Barry Constantine
Lect3..ppt - 09/12/04 CIS 4100 Systems Performance and Evaluation Lecture 3 by Zornitza Genova Prodanoff.
Transport Layer3-1 Chapter 3 outline r 3.1 Transport-layer services r 3.2 Multiplexing and demultiplexing r 3.3 Connectionless transport: UDP r 3.4 Principles.
Qian Zhang Department of Computer Science HKUST Advanced Topics in Next- Generation Wireless Networks Transport Protocols in Ad hoc Networks.
Top-Down Network Design Chapter Thirteen Optimizing Your Network Design Oppenheimer.
POSTECH DP&NM Lab. Internet Traffic Monitoring and Analysis: Methods and Applications (1) 2. Network Monitoring Metrics.
POSTECH DP&NM Lab. Internet Traffic Monitoring and Analysis: Methods and Applications (1) 4. Active Monitoring Techniques.
Principles of Congestion Control Congestion: informally: “too many sources sending too much data too fast for network to handle” different from flow control!
1 Transport Protocols (continued) Relates to Lab 5. UDP and TCP.
Chapter 12 Transmission Control Protocol (TCP)
© 2006 Cisco Systems, Inc. All rights reserved. Optimizing Converged Cisco Networks (ONT) Module 3: Introduction to IP QoS.
TCP Trunking: Design, Implementation and Performance H.T. Kung and S. Y. Wang.
Demystifying Quality of Service (QoS). Page 2 What Is Quality of Service?  Ability of a network to provide improved service to selected network traffic.
1 Traffic Management Benchmarking Framework IETF 85 Atlanta Barry Constantine Tim Copley Ram Krishnan.
Draft-ietf-ippm-tcp-throughput-tm-04.txt 1 TCP Throughput Testing Methodology IETF 78 Maastricht Reinhard Schrage Barry Constantine.
Transport Layer 3-1 Chapter 3 Transport Layer Computer Networking: A Top Down Approach 6 th edition Jim Kurose, Keith Ross Addison-Wesley March
CS640: Introduction to Computer Networks Aditya Akella Lecture 15 TCP – III Reliability and Implementation Issues.
Computer Networking Lecture 18 – More TCP & Congestion Control.
CS640: Introduction to Computer Networks Aditya Akella Lecture 15 TCP – III Reliability and Implementation Issues.
Transport Layer3-1 Chapter 3 outline r 3.1 Transport-layer services r 3.2 Multiplexing and demultiplexing r 3.3 Connectionless transport: UDP r 3.4 Principles.
Low Latency Adaptive Streaming over TCP Authors Ashvin Goel Charles Krasic Jonathan Walpole Presented By Sudeep Rege Sachin Edlabadkar.
1 Computer Networks Congestion Avoidance. 2 Recall TCP Sliding Window Operation.
Advance Computer Networks Lecture#09 & 10 Instructor: Engr. Muhammad Mateen Yaqoob.
Random Early Detection (RED) Router notifies source before congestion happens - just drop the packet (TCP will timeout and adjust its window) - could make.
TCP continued. Discussion – TCP Throughput TCP will most likely generate the saw tooth type of traffic. – A rough estimate is that the congestion window.
Midterm Review Chapter 1: Introduction Chapter 2: Application Layer
Transport Layer3-1 Chapter 3 outline r 3.1 Transport-layer services r 3.2 Multiplexing and demultiplexing r 3.3 Connectionless transport: UDP r 3.4 Principles.
Peer-to-Peer Networks 13 Internet – The Underlay Network
1 Traffic Management Benchmarking Framework IETF 90 Toronto draft-constantine-bmwg-traffic-management-04 Barry Constantine Tim.
Chapter 10 Congestion Control in Data Networks and Internets 1 Chapter 10 Congestion Control in Data Networks and Internets.
Bandwidth estimation: metrics, measurement techniques, and tools Presenter: Yuhang Wang.
Bandwidth Estimation: Metrics Measurement Techniques and Tools
Window Control Adjust transmission rate by changing Window Size
Instructor Materials Chapter 6: Quality of Service
TCP - Part II Relates to Lab 5. This is an extended module that covers TCP flow control, congestion control, and error control in TCP.
Topics discussed in this section:
Chapter 3 outline 3.1 transport-layer services
COMP 431 Internet Services & Protocols
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 6: Quality of Service Connecting Networks.
TCP - Part II Relates to Lab 5. This is an extended module that covers TCP flow control, congestion control, and error control in TCP.
Congestion Control, Quality of Service, & Internetworking
Congestion Control (from Chapter 05)
Congestion Control (from Chapter 05)
Transport Layer: Congestion Control
Chapter 3 outline 3.1 Transport-layer services
Presentation transcript:

draft-constantine-ippm-tcp-throughput-tm-02.txt 1 TCP Throughput Testing Methodology IETF 77 Anaheim Barry Constantine Reinhard Schrage

draft-constantine-ippm-tcp-throughput-tm-02.txt 2 OSI Model: Division of Responsibility IT department responsibility Network Provider’s responsibility Shared responsibility Transport 4 Application Presentation Session Network Datalink Physical HTTP, FTP, , etc. TCP IP Ethernet

draft-constantine-ippm-tcp-throughput-tm-02.txt 3 History: Provisioning of Managed Networks  Even though RFC2544 was originally intended to benchmark network equipment in a lab environment, network providers have used it to benchmark operational networks in order to provide Service Level Agreements (SLAs) to their business customers –It is difficult if not impossible, to extrapolate end-user application layer performance from RFC2544 results and the goal of RFC2544 was never intended to do so.  Network providers are coming to the realization that RFC2544 testing and TCP layer testing are required to more adequately ensure end-user satisfaction

draft-constantine-ippm-tcp-throughput-tm-00.txt 4 Network Provisioning– RFC 2544 Overview  Goal –Run a sequence of Layer 2 / 3 tests to verify the general performance of a circuit prior to “turning on” the end user service  Test method –Packet based end-end or looped-back  Test end-end network: –Throughput rate in frames/sec or % link utilization –Frame loss absolute or % –Delay/Latencyin ms or us –Back-to-Back in frames or time  Test parameters: –Packet size: 64, 128, 256, 512, 1024, 1280, 1518 bytes –Packet rate: 10, 20, 30, 40, 50, 60, 70, 80, 90, 100% of maximum rate –Burst: Time or number of packets

draft-constantine-ippm-tcp-throughput-tm-02.txt 5 The Challenge of Measuring TCP Throughput cwnd = current TCP estimation of available capacity in the network cwnd Time Slow Start Congestion Avoidance TCP Window halving upon loss Fast Retransmission cwnd=1 upon timeout Slow Start High ssthresh Adjusted ssthresh 1234 RTO Timer expiration The predicted Bulk Transport Capacity (BTC) occurs during the peak of the Congestion Avoidance phase In the draft, we call this the “maximum sustained TCP throughput”

draft-constantine-ippm-tcp-throughput-tm-02.txt 6 TCP Test Methodology Overview “0”. Run traditional RFC2544 to verify the integrity of the network before conducting TCP testing. 1.Path MTU Detection (per RFC4821) –Verify network MTU with active TCP segment size testing 2.Baseline Round-trip Delay and Bandwidth –Predict optimum TCP window size 3.Single TCP Connection Throughput Tests –Verify TCP window size predictions 4.MSS Throughput Tests –Verify expected throughout per various MSS sizes 5.Multiple TCP Connection Throughput Tests –Test for tail drop condition (default FIFO queuing), policing, etc. 6.TCP + Background UDP Prioritization Tests –Verify end-end network prioritization with stateful TCP traffic (multiple TCP connections)

draft-constantine-ippm-tcp-throughput-tm-02.txt 7 Step 1: Path MTU Detection (per RFC4821)  Verify network MTU with active TCP MSS testing using Packetization Layer Path MTU Discovery (PLPMTUD) –PLPMTUD is an extension to existing Path MTU Discovery methods described in RFC1191 and RFC1981  PLPMTUD uses TCP instead of ICMP to probe the network and discover the minimum MTU size supported –When ICMP messages are not supported by network devices, traditional PMTUD fails to properly identify the minimum MTU  The minimum MTU from this step is used as the basis for the MSS in subsequent tests

draft-constantine-ippm-tcp-throughput-tm-02.txt 8 Step 2: Baseline Round-trip Time and Bandwidth  Round-trip Time and Bandwidth capacity measurements provide estimates of the ideal TCP window size, which will be used in subsequent test steps. –These latency and bandwidth tests should be run long enough to characterize the performance of the network over the course of a meaningful time period. –The test tool must be capable of delivering the TCP capacity (PC tools for certain speeds, dedicated test tools for higher speeds such as 1G/10GigE) TCP In-flight Data Round Trip Time

draft-constantine-ippm-tcp-throughput-tm-02.txt 9 Bandwidth Measurement Interval Considerations  Bandwidth measurements can look very differently if the procedure to perform measurements is not clearly defined –Compare display of measurements of below simple network if done with different epoch intervals

draft-constantine-ippm-tcp-throughput-tm-02.txt 10 Step 3: Single TCP Connection Throughput Tests  With baseline measurements of round trip time and bandwidth, a series of single connection TCP throughput tests can be conducted to baseline the performance of the network against expectations –Provide chart of throughput, retransmissions, RTT over time and the concept of a “TCP Goodput” type metric –Provide table of “input” versus expected “output” guideline –And provide the formula(s) used to derive the TCP “output” guideline This is the BDP calculation with provisions for MSS size and L2+L3+L4 overhead

draft-constantine-ippm-tcp-throughput-tm-02.txt 11 Step 4 : MSS Throughput Tests  By varying the MSS size of the TCP connection(s), the ability of the network to sustain expected TCP throughput can be verified. –Similar to RFC2544 packet size tests, which determine a routing or switching device’s ability to handle loads in term of packets per second

draft-constantine-ippm-tcp-throughput-tm-02.txt 12 Step 5: Multiple TCP Connection Throughput Tests  Default router queuing (i.e. FIFO based) is inefficient for business critical applications. –Can cause TCP Tail Drop and Global Synchronization; from the user’s perspective, this condition causes significant performance degradation  By automating end-to-end testing with several (4 or more) simultaneous TCP sessions, detect non-optimized shaping / queuing in the network –i.e. an over-utilized link should employ a random early discard technique to avoid tail drop 45 Mb/sec Link, RED Queuing 45 Mb/sec Link, FIFO Queuing (Tail Drop)

draft-constantine-ippm-tcp-throughput-tm-02.txt 13 Step 6: TCP + Background UDP Prioritization Tests  Application traffic such as Citrix, Peoplesoft, etc. now require real-time performance to meet end-user response time expectations; there is a fine balance between application data traffic prioritization and VoIP, Video, etc. –Emulate bursty TCP traffic sessions (i.e. Citrix, HTTP, SMTP, etc.) with the proper CoS and QoS values at an average throughput rate and with peaks. –Emulate concurrent UDP sessions (i.e. VoIP G.711) with the proper CoS and QoS values TCP Session #1

draft-constantine-ippm-tcp-throughput-tm-02.txt 14 Next Steps for the TCP Testing Draft  Work with network providers to solicit feedback and to better align the testing methodology with practical application “in the field”  Define the “TCP Goodput” metric in the draft and provide recommended thresholds in various network topologies (metro fiber, DSL access, cable access, satellite, etc.)  Develop the MSS Throughput and Multiple Connection test sections to the next level of detail