1 / 14

Field Test Results of F-RTO

Field Test Results of F-RTO. Kazunori Yamamoto NTT DoCoMo yamamotokaz@nttdocomo.co.jp. William Gilliam Hewlett-Packard Company wag@cup.hp.com. Paris, August 2, 2005. Purpose. Validate the effectiveness of F-RTO with Eifel response through the field test Unnecessary retransmission

jael-little
Download Presentation

Field Test Results of F-RTO

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Field Test Results of F-RTO Kazunori Yamamoto NTT DoCoMo yamamotokaz@nttdocomo.co.jp William Gilliam Hewlett-Packard Company wag@cup.hp.com Paris, August 2, 2005

  2. Purpose • Validate the effectiveness of F-RTO with Eifel response through • the field test • Unnecessary retransmission • Throughput • Detection rate of Spurious Timeout (STO) • Implement F-RTO with Eifel response on the HP-UX 11i prototype. • Conduct tests in the commercial 3G mobile communication network.The service is known as “FOMA” and provided by NTT DoCoMo in Japan.

  3. 3G Mobile Communication Network (FOMA) Fixed point, Low speed With F-RTO UE 100Mbps Access Network Middle speed (Car) HP-UX11iPrototype Internet High speed (Bullet Train) Without F-RTO UE: User Equipment Location: Yokosuka Research park Test Environment

  4. Test Parameters

  5. Channel switch Passing many tunnels High frequency of handover Frequency of STO • STO occurred 0.26 times for 1MB data transfer on average. • STO occurred most frequently at high speed and fixed point with good wireless condition. (Once per 3MB data transfer)

  6. Reduction in Unnecessary Retransmission (1/3) Without F-RTO Unnecessary retransmission caused by STO 0.34% Unnecessary retransmission caused by STO Total Retransmitted Data 1.1% 31.2% 68.8% Original Data 98.9% Necessary retransmission 0.76% Necessary retransmission

  7. Reduction in Unnecessary Retransmission (2/3) With F-RTO Unnecessary retransmission caused by STO 0.06% Unnecessary retransmission caused by STO Total Retransmitted Data 0.85% 7.1% 92.9% Original Data 99.15% Necessary retransmission 0.79% Necessary retransmission

  8. 100 Without F-RTO With F-RTO 90 Reduction Rate in Unnecessary Retransmission (%) 80 70 Reduction in Unnecessary Retransmission (3/3) • Without F-RTO, unnecessarily retransmitted data due to STO accounts for 0.99% of the whole transmitted data at high speed. • With F-RTO, unnecessarily retransmitted data due to STO is reduced by over 80% in every environment.

  9. Throughput With F-RTO, the throughput is increased by 6% for the connections that suffered STO more than once and transferred at least 2MB data. 277.7 264 6% Up

  10. Detection Rate of F-RTO Duplicate ACKs were received after STO. Receive window restricted new data transmission after STO. 0.7% 0.3% STO not detected 29% STO detected 71% 28% STO occurred after FIN sent.

  11. Experimental Environment After reaching MaxWin Congestion Window Size (KB) Mid. stage of slow start Early stage of slow start

  12. Without F-RTO With F-RTO Experimental Results • The effects of F-RTO, in terms of unnecessary retransmission and throughput, become large with higher wireless speed. • Throughput is increased by a maximum of 262%. Unnecessary Retransmission Throughput 500 12 300 9.7 10.1 10 377 10 400 7.3 8 200 300 5.2 227 Increase Rate in Throughput Throughput (Mbps) Unnecessary Retransmission (segment) 6 200 4 100 2.7 100 2 5 1 1 1 0 0 0 Early Mid. MaxWin Early Mid. MaxWin

  13. Conclusions • Unnecessary Retransmission • Without F-RTO, unnecessary retransmission due to STO accounts for 31% of the totalretransmission. • F-RTO can reduce unnecessary retransmission to 7%. • Throughput • F-RTO can increase throughput by 6% in the 3G mobile communications network. • F-RTO can increase throughput by 262% in higher speed mobile communication network (HSDPA). • Detection Rate F-RTO can detect 71% of STO. Good enough to become Proposed Standard?

  14. Papers • K. Yamamoto, et. al., “Effects of F-RTO and Eifel Response Algorithms for W-CDMA and HSDPA networks,” Wireless Personal Multimedia Communications(WPMC)’05, Sept. 2005. • A. Hokamura, et. al., “Performance Evaluation of F-RTO and Eifel Response Algorithms over W-CDMA packet network,” Wireless Personal Multimedia Communications(WPMC)’05, Sept. 2005.

More Related