Abstract

A congestion control scheme called dual-layer congestion control (DLCC) is proposed for use when transporting Internet traffic over optical-packet-switched networks. It further reduces the core optical buffering requirement over existing proposals; indeed each optical core switch is assumed in the modeling work to have a shared optical buffering capacity of only 20 optical packets for all ports. Furthermore, it does not depend for its operation on having a certain number of Transmission Control Protocol (TCP) flows carried over each link. The scheme is designed to operate in conjunction with an edge-smoothing algorithm that segments IP datagrams into fixed-length optical slots to be carried by the core. It expedites the response of TCP to congestion in the optical core network, both by reducing the rate of packet transmission over the optical packet core and by throttling TCP sources via the transmission of additional triple duplicate ACK segments. Packet loss performance and edge-buffering capacity requirements are evaluated through mathematical analysis, showing that the packet loss rate can be decreased through the use of DLCC by a factor of up to six times and also showing that electronic edge-buffering requirements are reduced through the use of DLCC. Furthermore, simulation modeling shows that DLCC yields a TCP goodput improvement of between 2 and 10 times, depending on the volume of background User Data Protocol (UDP) traffic and the round-trip time. This demonstrates that DLCC is viable and enhances network performance.

© 2009 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. D. K. Hunter, M. C. Chia, I. Andonovic, “Buffering in optical packet switches,” J. Lightwave Technol., vol. 16, pp. 2081–2094, 1998.
    [CrossRef]
  2. R. S. Tucker, “The role of optics and electronics in high-capacity routers,” J. Lightwave Technol., vol. 24, pp. 4655–4673, 2006.
    [CrossRef]
  3. R. S. Tucker, P.-C. Ku, C. J. Chang-Hasnain, “Slow-light optical buffers: capabilities and fundamental limitations,” J. Lightwave Technol., vol. 23, pp. 4046–4066, 2005.
    [CrossRef]
  4. G. Appenzeller, I. Keslassy, K. McKeown, “Sizing router buffers,” ACM SIGCOMM Comput. Rev., vol. 34, no. 4, pp. 281–292, Oct. 2004.
  5. Z. Lu, D. K. Hunter, I. D. Henning, “Contention reduction in core optical packet switches through electronic traffic smoothing and scheduling at the network edge,” J. Lightwave Technol., vol. 24, pp. 4828–4837, 2006.
    [CrossRef]
  6. O. Alparslan, S. Arakawa, M. Murata, “Rate-based pacing for small buffered optical packet-switched networks,” J. Opt. Netw., vol. 6, pp. 1116–1128, 2007.
    [CrossRef]
  7. E. Hashem, “Analysis of random drop for gateway congestion control,” Laboratory for Computer Science, MIT, Cambridge, MA, Rep. LCS TR-465, 1989.
  8. S. Floyd, V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, pp. 397–413, 1993.
    [CrossRef]
  9. S. Floyd, “TCP and explicit congestion notification,” Comput. Commun. Rev., vol. 24, pp. 10–23, 1994.
    [CrossRef]
  10. F. Xue, S. Yoo, “TCP-aware active congestion control in optical packet-switched networks,” in Optical Fiber Communication Conf., 2003, paper MF108.
  11. S. Floyd, “A report on recent developments in TCP congestion control,” IEEE Commun. Mag., vol. 39, no. 4, pp. 84–90, Feb. 2001.
    [CrossRef]
  12. K. Claffy, G. Miller, K. Thompson, “The nature of the beast: recent traffic measurements from an Internet backbone,” in Proc. INET, Geneva, Switzerland, 1998, paper 473.
  13. C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
    [CrossRef]
  14. M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.
  15. S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.
  16. J. He, S.-H. G. Chan, “TCP and UDP performance for Internet over optical packet-switched networks,” in IEEE Int. Conf. Communications, 2003, pp. 1350–1354.
  17. K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
    [CrossRef]
  18. X. Yu, Y. Chen, C. Qiao, “Performance evaluation of optical burst switching with assembled burst traffic input,” in IEEE Global Telecommunications Conf., 2002, pp. 2318–2322.
  19. X. Yu, J. Li, X. Cao, Y. Chen, C. Qiao, “Traffic statistics and performance evaluation in optical burst switched networks,” J. Lightwave Technol., vol. 22, pp. 2722–2738, 2004.
    [CrossRef]
  20. M. Allman, V. Paxson, W. Stevens, “TCP congestion control,” RFC 2581, IETF, 1999.
  21. S. Floyd, “Congestion control principles,” RFC 2914, IETF, 2000.
  22. G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
    [CrossRef]
  23. D. Wischik, N. McKeown, “Part I: Buffer sizes for core routers,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 75–78, July 2005.
    [CrossRef]
  24. D. Wischik, “Buffer requirements for high-speed routers,” in 31st European Conf. Optical Communication, 2005, pp. 23–26.
  25. OPNET network simulator, http://www.opnet.com.

2007

2006

2005

R. S. Tucker, P.-C. Ku, C. J. Chang-Hasnain, “Slow-light optical buffers: capabilities and fundamental limitations,” J. Lightwave Technol., vol. 23, pp. 4046–4066, 2005.
[CrossRef]

G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
[CrossRef]

D. Wischik, N. McKeown, “Part I: Buffer sizes for core routers,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 75–78, July 2005.
[CrossRef]

2004

2003

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

2001

S. Floyd, “A report on recent developments in TCP congestion control,” IEEE Commun. Mag., vol. 39, no. 4, pp. 84–90, Feb. 2001.
[CrossRef]

1998

1997

K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
[CrossRef]

1994

S. Floyd, “TCP and explicit congestion notification,” Comput. Commun. Rev., vol. 24, pp. 10–23, 1994.
[CrossRef]

1993

S. Floyd, V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, pp. 397–413, 1993.
[CrossRef]

Allman, M.

M. Allman, V. Paxson, W. Stevens, “TCP congestion control,” RFC 2581, IETF, 1999.

Alparslan, O.

Andonovic, I.

Appenzeller, G.

G. Appenzeller, I. Keslassy, K. McKeown, “Sizing router buffers,” ACM SIGCOMM Comput. Rev., vol. 34, no. 4, pp. 281–292, Oct. 2004.

Arakawa, S.

Bhattacharyya, S.

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

Cao, X.

Chan, S.-H. G.

J. He, S.-H. G. Chan, “TCP and UDP performance for Internet over optical packet-switched networks,” in IEEE Int. Conf. Communications, 2003, pp. 1350–1354.

Chang-Hasnain, C. J.

Chen, Y.

X. Yu, J. Li, X. Cao, Y. Chen, C. Qiao, “Traffic statistics and performance evaluation in optical burst switched networks,” J. Lightwave Technol., vol. 22, pp. 2722–2738, 2004.
[CrossRef]

X. Yu, Y. Chen, C. Qiao, “Performance evaluation of optical burst switching with assembled burst traffic input,” in IEEE Global Telecommunications Conf., 2002, pp. 2318–2322.

Chia, M. C.

Claffy, K.

K. Claffy, G. Miller, K. Thompson, “The nature of the beast: recent traffic measurements from an Internet backbone,” in Proc. INET, Geneva, Switzerland, 1998, paper 473.

M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.

Cotton, C.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Diot, C.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

Floyd, S.

S. Floyd, “A report on recent developments in TCP congestion control,” IEEE Commun. Mag., vol. 39, no. 4, pp. 84–90, Feb. 2001.
[CrossRef]

S. Floyd, “TCP and explicit congestion notification,” Comput. Commun. Rev., vol. 24, pp. 10–23, 1994.
[CrossRef]

S. Floyd, V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, pp. 397–413, 1993.
[CrossRef]

S. Floyd, “Congestion control principles,” RFC 2914, IETF, 2000.

Fomenkov, M.

M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.

Fraleigh, C.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Hashem, E.

E. Hashem, “Analysis of random drop for gateway congestion control,” Laboratory for Computer Science, MIT, Cambridge, MA, Rep. LCS TR-465, 1989.

He, J.

J. He, S.-H. G. Chan, “TCP and UDP performance for Internet over optical packet-switched networks,” in IEEE Int. Conf. Communications, 2003, pp. 1350–1354.

Henning, I. D.

Hunter, D. K.

Iyer, S.

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

Jacobson, V.

S. Floyd, V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, pp. 397–413, 1993.
[CrossRef]

Keslassy, I.

G. Appenzeller, I. Keslassy, K. McKeown, “Sizing router buffers,” ACM SIGCOMM Comput. Rev., vol. 34, no. 4, pp. 281–292, Oct. 2004.

Keys, K.

M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.

Khan, M.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Ku, P.-C.

Li, J.

Lu, Z.

Lyles, B.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

McKeown, K.

G. Appenzeller, I. Keslassy, K. McKeown, “Sizing router buffers,” ACM SIGCOMM Comput. Rev., vol. 34, no. 4, pp. 281–292, Oct. 2004.

McKeown, N.

D. Wischik, N. McKeown, “Part I: Buffer sizes for core routers,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 75–78, July 2005.
[CrossRef]

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

Miller, G.

K. Claffy, G. Miller, K. Thompson, “The nature of the beast: recent traffic measurements from an Internet backbone,” in Proc. INET, Geneva, Switzerland, 1998, paper 473.

Miller, G. J.

K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
[CrossRef]

Moll, D.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Moon, S.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Moore, D.

M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.

Murata, M.

Paxson, V.

M. Allman, V. Paxson, W. Stevens, “TCP congestion control,” RFC 2581, IETF, 1999.

Qiao, C.

X. Yu, J. Li, X. Cao, Y. Chen, C. Qiao, “Traffic statistics and performance evaluation in optical burst switched networks,” J. Lightwave Technol., vol. 22, pp. 2722–2738, 2004.
[CrossRef]

X. Yu, Y. Chen, C. Qiao, “Performance evaluation of optical burst switching with assembled burst traffic input,” in IEEE Global Telecommunications Conf., 2002, pp. 2318–2322.

Raina, G.

G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
[CrossRef]

Rockell, R.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Seely, T.

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

Stevens, W.

M. Allman, V. Paxson, W. Stevens, “TCP congestion control,” RFC 2581, IETF, 1999.

Taft, N.

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

Thompson, K.

K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
[CrossRef]

K. Claffy, G. Miller, K. Thompson, “The nature of the beast: recent traffic measurements from an Internet backbone,” in Proc. INET, Geneva, Switzerland, 1998, paper 473.

Towsley, D.

G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
[CrossRef]

Tucker, R. S.

Wilder, R.

K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
[CrossRef]

Wischik, D.

D. Wischik, N. McKeown, “Part I: Buffer sizes for core routers,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 75–78, July 2005.
[CrossRef]

G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
[CrossRef]

D. Wischik, “Buffer requirements for high-speed routers,” in 31st European Conf. Optical Communication, 2005, pp. 23–26.

Xue, F.

F. Xue, S. Yoo, “TCP-aware active congestion control in optical packet-switched networks,” in Optical Fiber Communication Conf., 2003, paper MF108.

Yoo, S.

F. Xue, S. Yoo, “TCP-aware active congestion control in optical packet-switched networks,” in Optical Fiber Communication Conf., 2003, paper MF108.

Yu, X.

X. Yu, J. Li, X. Cao, Y. Chen, C. Qiao, “Traffic statistics and performance evaluation in optical burst switched networks,” J. Lightwave Technol., vol. 22, pp. 2722–2738, 2004.
[CrossRef]

X. Yu, Y. Chen, C. Qiao, “Performance evaluation of optical burst switching with assembled burst traffic input,” in IEEE Global Telecommunications Conf., 2002, pp. 2318–2322.

ACM SIGCOMM Comput. Commun. Rev.

G. Raina, D. Towsley, D. Wischik, “Part II: Control theory for buffer sizing,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 79–82, July 2005.
[CrossRef]

D. Wischik, N. McKeown, “Part I: Buffer sizes for core routers,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 75–78, July 2005.
[CrossRef]

Comput. Commun. Rev.

S. Floyd, “TCP and explicit congestion notification,” Comput. Commun. Rev., vol. 24, pp. 10–23, 1994.
[CrossRef]

IEEE Commun. Mag.

S. Floyd, “A report on recent developments in TCP congestion control,” IEEE Commun. Mag., vol. 39, no. 4, pp. 84–90, Feb. 2001.
[CrossRef]

IEEE Network

C. Fraleigh, S. Moon, B. Lyles, C. Cotton, M. Khan, D. Moll, R. Rockell, T. Seely, C. Diot, “Packet-level traffic measurements from the Sprint IP backbone,” IEEE Network, vol. 17, pp. 6–16, 2003.
[CrossRef]

K. Thompson, G. J. Miller, R. Wilder, “Wide-area Internet traffic patterns and characteristics,” IEEE Network, vol. 11, pp. 10–23, 1997.
[CrossRef]

IEEE/ACM Trans. Netw.

S. Floyd, V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, pp. 397–413, 1993.
[CrossRef]

J. Lightwave Technol.

J. Opt. Netw.

Other

K. Claffy, G. Miller, K. Thompson, “The nature of the beast: recent traffic measurements from an Internet backbone,” in Proc. INET, Geneva, Switzerland, 1998, paper 473.

E. Hashem, “Analysis of random drop for gateway congestion control,” Laboratory for Computer Science, MIT, Cambridge, MA, Rep. LCS TR-465, 1989.

F. Xue, S. Yoo, “TCP-aware active congestion control in optical packet-switched networks,” in Optical Fiber Communication Conf., 2003, paper MF108.

M. Fomenkov, K. Keys, D. Moore, K. Claffy, “Longitudinal study of Internet traffic in 1998–2003,” in Proc. Winter Int. Symp. Information and Communication Technologies, Cancun, Mexico, 2004, pp. 1–6.

S. Iyer, S. Bhattacharyya, N. Taft, N. McKeown, C. Diot, “An approach to alleviate link overload as observed on an IP backbone,” in 22nd Annu. Joint Conf. IEEE Computer and Communications Societies, 2003, pp. 406–416.

J. He, S.-H. G. Chan, “TCP and UDP performance for Internet over optical packet-switched networks,” in IEEE Int. Conf. Communications, 2003, pp. 1350–1354.

X. Yu, Y. Chen, C. Qiao, “Performance evaluation of optical burst switching with assembled burst traffic input,” in IEEE Global Telecommunications Conf., 2002, pp. 2318–2322.

G. Appenzeller, I. Keslassy, K. McKeown, “Sizing router buffers,” ACM SIGCOMM Comput. Rev., vol. 34, no. 4, pp. 281–292, Oct. 2004.

M. Allman, V. Paxson, W. Stevens, “TCP congestion control,” RFC 2581, IETF, 1999.

S. Floyd, “Congestion control principles,” RFC 2914, IETF, 2000.

D. Wischik, “Buffer requirements for high-speed routers,” in 31st European Conf. Optical Communication, 2005, pp. 23–26.

OPNET network simulator, http://www.opnet.com.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (12)

Fig. 1
Fig. 1

Timeline illustration of DLCC, showing at the top the scenario used for modeling, which consists of a TCP transmitter, edge routers, a core switch experiencing heavy congestion and exhibiting packet loss, and the TCP receiver. T 0 , T 1 , T 2 , and T 3 are the propagation delays of each network segment. The following events take place with DLCC in order to expedite TCP’s response to congestion: 1. A core optical packet switch experiences packet loss and determines that this is due to heavy congestion. 2. The edge router receives a congestion notification from the core switch and reduces its transmission rate of smoothed optical packets into the core. 3. Four artificial ACK packets arrive at the TCP transmitter from the edge router, and TCP then halves its congestion window through the fast recovery feature. 4. The TCP receiver sends duplicate ACKs because a segment has been lost. 5. These duplicate ACKs are discarded by the edge router because it sent the duplicate ACKs to the TCP transmitter earlier.

Fig. 2
Fig. 2

Mean loss rate during heavy congestion for a single TCP flow and for multiple synchronized TCP flows.

Fig. 3
Fig. 3

Mean loss rate during heavy congestion for 10,000 nonsynchronized TCP flows.

Fig. 4
Fig. 4

Mean total number of lost optical packets over one period for 10,000 nonsynchronized TCP flows.

Fig. 5
Fig. 5

Mean electronic edge buffer requirement in number of optical packets for a single TCP flow and for multiple synchronized TCP flows. With “1 control,” the OPS ACKs scheme is not implemented, whereas “2 controls” refers to the use of full dual-layer control.

Fig. 6
Fig. 6

Mean electronic edge buffer requirement in number of optical packets for 10,000 nonsynchronized TCP flows. With “1 control,” the OPS ACKs scheme is not implemented, whereas “2 controls” refers to the use of full dual-layer control.

Fig. 7
Fig. 7

Mean loss of the scenarios with control (with DLCC) during one congestion period, expressed as number of optical packets against number of nonsynchronized TCP flows.

Fig. 8
Fig. 8

Goodput comparison with varying end-to-end round-trip time, assuming that a maximum of 30% of the transmission capacity can be used by UDP traffic.

Fig. 9
Fig. 9

Extra edge buffer capacity with varying maximum UDP capacity ratio, assuming a round-trip time of 500   ms .

Fig. 10
Fig. 10

Extra edge buffer capacity with varying end-to-end round-trip time, assuming that a maximum of 30% of the transmission capacity can be used by UDP traffic.

Fig. 11
Fig. 11

Goodput comparison with varying maximum UDP capacity ratio, assuming a round-trip time of 500   ms .

Fig. 12
Fig. 12

Loss rate comparison with varying end-to-end round-trip time, assuming that a maximum of 30% of the transmission capacity can be used by UDP traffic.

Equations (27)

Equations on this page are rendered with MathJax. Learn more.

R i + 1 = ν R i + ( 1 ν ) ( U i + B τ ) .
P L = a i P ( x = a i B i ) L i 2 T C = a i P ( B i x = a i ) P ( x = a i ) L i 2 T C .
P ( B i x = a i ) = P ( a i + 1 > a i ) P ( a i + 1 > C W a i r 2 T × S + H S ) = P ( a i + 1 > a i ) × P ( a i + 1 > X ) .
P ( a i + 1 a i > 0 ) = P ( a i + 1 > a i ) = C R a i C R a i ( a i ) = C R a i C R .
P ( a i + 1 > a i ) P ( a i + 1 > X ) = P ( a i + 1 > max ( a i , X ) ) = X = 0 a i a i + 1 = a i C R d a i + 1 d X + X = a i C R a i + 1 = X C R d a i + 1 d X X = 0 C R a i + 1 = a i C R d a i + 1 d X .
W a i r = ( T i UDP W k 2 T S ) mod ( 2 T M ) 2 ( 2 T M ) 2 W a i p = { ( T i UDP W k 2 T S ) mod [ 2 T 2 ( C a i ) S + H ] } S T .
L i = 1 M k = 1 M a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X ) × 2 T ,
L i = 1 M k = 1 M a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X ) × 2 T 1 .
B ¯ 1 c = a i P ( x = a i ) 1 M k = 1 M a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X ) × ( 2 T 2 T 1 ) .
B ¯ 2 c = a i P ( x = a i ) 1 M k = 1 M a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X ) × 2 T 0 .
W a i p N = 2 T ( C a i ) S N ( S + H ) ,
M N = 2 T ( C a i ) N ( S + H ) ,
X N = C N W a i r N ( S + H ) ( 2 T S ) ,
W k = k S .
L 1 N = 1 M N k = 1 M N a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X N ) × 2 T .
L 1 N = 1 M N k = 1 M N a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 X N ) × 2 T 1 .
B ¯ 1 c = a i P ( x = a i ) 1 M N k = 1 M N a i + 1 = a i C R P ( x = a i + 1 ) ( a i + 1 X N ) ( 2 T 2 T 1 ) .
B ¯ 2 c = a i P ( x = a i ) 1 M N k = 1 M N a i + 1 = a i C R P ( x = a i + 1 ) ( a i + 1 X N ) 2 T 0 .
r i e = a i + 1 = a i C R P ( x = a i + 1 ) × ( a i + 1 a i ) .
2 T + ( K 1 ) ( M N 2 T 2 ) N .
L i = r i e 2 T + r i e 2 ( K 1 ) ( M N 2 T 2 ) N .
P ( a i + 1 > a i ) = C R a i C R .
L ¯ = a i P ( x = a i ) L i .
L i = r i e 2 T 1 .
L ¯ = a i P ( x = a i ) r i e 2 T 1 .
B ¯ 1 c = a i P ( x = a i ) r i e ( 2 T 2 T 1 ) .
B ¯ 2 c = a i P ( x = a i ) r i e 2 T 0 .