Presentation is loading. Please wait.

Presentation is loading. Please wait.

Effects of Advertised Receive Buffer Size and Timer Granularity on TCP Performance in a LEO Satellite Network.

Similar presentations


Presentation on theme: "Effects of Advertised Receive Buffer Size and Timer Granularity on TCP Performance in a LEO Satellite Network."— Presentation transcript:

1 Effects of Advertised Receive Buffer Size and Timer Granularity on TCP Performance in a LEO Satellite Network

2 Marc Emmelmann, emmelmann@ieee.org Fraunhofer Institute for Open Communication Systems IEEE Globecom 2002 Taipei, Taiwan 17-21November 2002

3 Page 3 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Outline Introduction-Project Framework -Target System Parameter -TCP Basics Simulation Environment-Used Simulation Models -Simulation Parameters Simulation Results-Effects of Buffer Size & Timer Granularity -Effects of Bit Error Rates -Overall TCP Performance Conclusion & Outlook

4 Page 4 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Project Framework Partner-Tesat-Spacecom -German Aerospace Agency DLR -Fraunhofer FOKUS Satellite Network-Support of multimedia applications -LEO orbits -ATM based Project AspectsTarget System Design Antennas Demonstrator TCP Simulation (Reno)

5 Page 5 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Target System Design System Parameter Satellite Orbit-72 satellites (12 * 6) -Orbit altitude: 1350 km -Inclination 47 deg. -ISLs: on-board processing and switching Bit Rates-Uplink: up to 2.048 Mbit/s -Downlink: up to 32.768 Mbit/s -16 kbit/s granularity Bit Error Rates-Target BER < 2*10e-8 -Shadowing -Adaptive FEC Data Transfer-Usage of TCP Reno

6 Page 6 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan TCP Basics Provides-reliable end-to-end transfer Retransmissions caused by-expiration of retransmission timeout RTO -duplicate ACKs (Fast Retransmission) Calculates the RTO-upon smoothed RTT -timer granularity G acts as low pass filter RTO-may cause premature retransmissions (RTO > RTT)

7 Page 7 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Model Delay--analytically gained -simulation based Adapted point-2-point link-BER -bandwidth, delay -queuing behavior FTP Transfer-File Size

8 Page 8 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Parameter Lower Bound Of RTT in msNone100200Default Timer Granularity G G=1 ms27 G=50 ms38 G=100 ms49 G=250 ms510 G=500 ms61112 Minimal G to avoid false retransmissions (varying)1 Given: -TCP Flavor -Buffer Size -BER More than 140 experiments

9 Page 9 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Results Effects of Buffer Size & Timer Granularity Advertised Receive Buffer-Larger than BDP -Less than BDP -Equal to MSS Timer Granularity-May cause false retransmissions Advertised Receive Buffer 64 kB >> BDP 6.6 kB = BDP 3.3 kB 1/2 BDP 1.46 kB MSS Min. Timer Granularity (G)1 ms 4 ms26 ms (250 ms)

10 Page 10 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Results Effects of Bit Error Rates BER = 2*10e-8-Overlapping with BER = 0 -No needless retransmissions

11 Page 11 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Results Effects of Bit Error Rates (cont.) BER = 2 * 10e-5-No resemblance with variable propagation delay -Prevention of false retransmissions

12 Page 12 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Results TCP Performance Degradation of TCP throughput BER = 2* 10e-8

13 Page 13 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Results TCP Performance Degradation of TCP throughput BER 2*10e-5

14 Page 14 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Conclusion & Outlook TCP-Sophisticated protocol -Minor influence of hand-tuned implementations -TCP flavor dominates Demonstrator-Run experiments over “real” MAC -Combine simulation models and demonstrator Further Information-http://www.fokus.fhg.de/cats/satellite -emmelmann@ieee.org

15 Page 15 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Demonstrator Modular Design-Support of multiple MAC- and Signaling Layers -COTS equipment -Re-Use of software components Standard Environment-Smooth workflow from simulation to implementation -Integrated demonstration and development environment

16 Page 16 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Demonstrator Simulation Scenario & Architecture two terminal in different footprints satellite satellite channel application terminal satellite channel emulation control station terminal ISL SDLC TDLC Architecture DLC system separated from the terminal to be independent from OS Satellite Channel emulated in software based on precalculated tables Control Station manages the environment and the satellite

17 Page 17 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Environment Simulation Workflow Definition of satellite network-Orbit altitude -Orbit type Definition of communication-Bandwidth network aspects-Bit Error Rates -Internet Protocol Suite AER Data Satellite Path Description *.c and *.h Files for pipeline stages Parser Programs p2p-link w/ variable delay STK Opnet

18 Page 18 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Simulation Environment Simulation Model Verification Simulation Models for Research related Experiments Simulation Models for System Verification

19 Page 19 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan ATM-Sat Satellite Delays

20 Page 20 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Verification of Network Model PTP-Duplex-Link: - delay - BW - BER

21 Page 21 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan Network Model Verification Verification Measurements Delay-Expected values -Consider queuing delay Bit rate-Upper Limit

22 Page 22 emmelmnn@ieee.orgemmelmnn@ieee.org IEEE Globecom ‘02, Nov 17-21 2002, Taipei, Taiwan ATM-SAT FEC schemes -Systems with high link margins are wasting transmission capacity most of the time! -Solution: optimized MAC protocol, which saves transmission capacity by means of adaptive controlling the cell error rate, e.g. by: - adaptive modulation - adaptive coding QPSK,RS(65,53),Rate 1/2 BlockTurbo Code 2.46 QPSK,RS(65,53) 1.23 Burst- length QPSK 1 16 QAM 0.5 E=90°E=40°E=20° Example for CER th =10 -6 : 0 mm/h 2 mm/h 16 mm/h Uplink (30 GHz) 4 mm/h


Download ppt "Effects of Advertised Receive Buffer Size and Timer Granularity on TCP Performance in a LEO Satellite Network."

Similar presentations


Ads by Google