Method and apparatus for the assessment and optimization of network traffic

7,336,613

10/070515

October 17, 2001

A system for the assessment of network performance criteria, and applying this criteria to the classification of network addresses into appropriate ranges, using these ranges to consolidate performance measurements for the associated addresses, and applying these metrics toward the optimization of the network towards performance or policy objectives.

Lloyd, Michael A. (Belmont, CA, US); Baldonado, Omar C. (Palo Alto, CA, US); Finn, Sean P. (Belmont, CA, US); Karam, Mansour J. (Mountain View, CA, US); McGuire, James G. (San Francisco, CA, US); Madan, Herbert S. (Tiburon, CA, US)

Avaya Technology Corp. (Basking Ridge, NJ, US)

1. A method of optimizing traffic in an internetwork, the method comprising: selecting a subset of flows in the internetwork for monitoring, wherein the subset of flows includes one of one flow, some flows, and all flows; measuring performance characteristics of the subset of flows in at least a portion of the internetwork, the performance characteristics including one or more of: a plurality of one or more round trip measurements for each of the subset of flows and a plurality of one or more one-way measurements for each of the subset of flows, the measuring performance characteristics including: launching a first plurality of one or more packets, and measuring the first plurality of one or more packets; measuring a second plurality of one or more packets in the internetwork, wherein the second plurality of one or more packets were already launched; monitoring a first plurality of one or more flows in the internetwork, duplicating one or more packets from the plurality of one or more flows, and measuring the one or more duplicated packets; serving as a proxy hop for a second plurality of one or more flows, and measuring the second plurality of one or more flows; and encoding material within a third plurality of one or more flows, causing a fourth plurality of one or more flows to be generated, wherein the third plurality of one or more flows traverses a first path of the internetwork, and the fourth plurality of one or more flows traverses a second path of the internetwork, wherein at least a portion of the first path of the internetwork and at least a portion of the second path of the internetwork are equal, and measuring a subset of the fourth plurality of one or more flows; calculating at least one performance metric for the subset of flows in the at least a portion of the internetwork, the at least one performance metric at least partly determined from the measured performance characteristics; and in response to calculating the at least one performance metric, rearchitecting the internetwork to optimize one or more of the least one performance metric using peer-to-peer communications, rearchitecting the internetwork including at least one of: altering a plurality of one or more routing tables in the internetwork, wherein the plurality of one or more routing tables include at least one of: network-layer routing tables, layer 3 routing tables, IP routing tables, layer 2 forwarding tables, and MPLS forwarding tables; redirecting the subset of flows to a second internetwork coupled to the internetwork at one or more Points of Presence; and affecting forwarding decisions of the subset of flows, by imposing one or more of: NAT, GRE, and tunneling techniques other than GRE.

2. The method of claim 1 , wherein the plurality of one or more routing tables includes network-layer routing tables.

3. The method of claim 1 , wherein the plurality of one or more routing tables includes layer 3 routing tables.

4. The method of claim 1 , wherein the plurality of one or more routing tables includes IP routing tables.

5. The method of claim 1 , wherein the plurality of one or more routing tables includes layer 2 routing tables.

6. The method of claim 1 , wherein the plurality of one or more routing tables includes MPLS forwarding tables.

7. The method of claim 1 , wherein the performance characteristics includes the plurality of one or more round trip measurements.

8. The method of claim 1 , wherein the performance characteristics includes the plurality of one or more one-way measurements.

9. The method of claim 1 , wherein the one or more of the at least one performance metric to be optimized quantifies path performance of the subset of flows.

10. The method of claim 9 , wherein the one or more of the at least one performance metric characterizes a quality of a network application.

11. The method of claim 10 , wherein the application includes http.

12. The method of claim 10 , wherein the application includes ftp.

13. The method of claim 10 , wherein the application includes telnet.

14. The method of claim 10 , wherein the application includes voice.

15. The method of claim 10 , wherein the application include video.

16. The method of claim 1 , wherein one or more of the at least one performance metric quantifies one or more of: overall usage of the at least the portion of the internetwork, absolute individual network link usage in the at least the portion of the internetwork, relative individual network link usage in the at least the portion of the internetwork, and link usage cost in the at least the portion of the internetwork.

17. The method of claim 1 , wherein the subset of flows include at least a first sub-plurality of one or more flows and a second sub-plurality of one or more flows, and the first sub-plurality of one or more flows uses at least a first measured performance characteristic, and the second sub-plurality of one or more flows uses at least a second measured performance characteristic.

18. The method of claim 1 , wherein the one or more of the at least one performance metric quantifies geographic distance covered by the subset of flows in the internetwork.

19. The method of claim 1 , wherein the selection of the subset of flows is user-based in the at least the portion of the internetwork.

20. The method of claim 1 , wherein the internetwork comprises an autonomous sub-system of a larger network.

21. The method of claim 20 , wherein the larger network is the Internet.

22. The method of claim 20 , wherein the internetwork is a BGP autonomous system.

23. The method of claim 1 , wherein the internetwork comprises an autonomous system of a larger network.

24. The method of claim 23 , wherein the larger network is the Internet.

25. The method of claim 23 , wherein the internetwork is a BGP autonomous system.

26. The method of claim 1 , wherein the internetwork comprises an overlay network.

27. The method of claim 1 , wherein the internetwork comprises a plurality of one of more networks, the plurality of one or more networks coupled together, wherein the plurality of one or more networks include one or more of local-area networks, metropolitan-area networks, and wide-area networks.

28. The method of claim 1 , wherein the measuring performance characteristics includes launching a first plurality of one or more packets, and measuring the first plurality of one or more packets.

29. The method of claim 28 , wherein the measuring performance characteristics includes launching a first plurality of one or more round-trip packets, and measuring the first plurality of one or more round-trip packets, such that the plurality of one or more round trip measurements include round trip measurements for the launched packets.

30. The method of claim 29 , wherein the first plurality of one or more round-trip packets include traceroute ICMP packets.

31. The method of claim 29 , wherein the first plurality of one or more round-trip packets include ping ICMP packets.

32. The method of claim 29 , wherein the first plurality of one or more round-trip packets include telnet packets.

33. The method of claim 29 , wherein the first plurality of one or more round-trip packets include TCP packets from an empty TCP transaction.

34. The method of claim 29 , wherein the first plurality of one or more round-trip packets include http packets.

35. The method of claim 28 , wherein the measuring performance characteristics includes launching a first plurality of one or more one-way packets, and measuring the first plurality of one or more one-way packets, such that the plurality of one or more one-way measurements include one-way measurements for the launched packets.

36. The method of 1 , wherein the measuring performance characteristics includes measuring a second plurality of one or more packets in the internetwork, wherein the second plurality of one or more packets were already launched.

37. The method of claim 1 , wherein round-trip measurements include one or more of round-trip delay, round-trip jitter, round-trip loss, round-trip available bandwidth, and round-trip total bandwidth.

38. The method of claim 37 , wherein round-trip measurements include round-trip delay.

39. The method of claim 37 , wherein round-trip measurements include round-trip jitter.

40. The method of claim 37 , wherein round-trip measurements include round-trip loss.

41. The method of claim 37 , wherein round-trip measurements include round-trip available bandwidth.

42. The method of claim 37 , wherein round-trip measurements include round-trip total bandwidth.

43. The method of claim 1 , wherein one-way measurements include one or more of one-way delay, one-way jitter, one-way loss, one-way available bandwidth, and one-way total bandwidth.

44. The method of claim 43 , wherein one-way measurements include one-way delay.

45. The method of claim 43 , wherein one-way measurements include one-way jitter.

46. The method of claim 43 , wherein one-way measurements include one-way loss.

47. The method of claim 43 , wherein one-way measurements include one-way available bandwidth.

48. The method of claim 43 , wherein one-way measurements include one-way total bandwidth.

49. The method of claim 1 , wherein rearchitecting the internetwork includes altering a plurality of one or more routing tables in the internetwork.

50. The method of claim 49 , wherein the altering of the plurality of one or more routing tables is applied automatically.

51. The method of claim 49 , wherein the altering of the plurality of one or more routing tables includes configuring a plurality of one or more routers, wherein the configuring the plurality of one or more routers statically alter the routing of flows.

52. The method of claim 51 , wherein the configuring a plurality of one or more routers includes route maps.

53. The method of claim 51 , wherein the configuring a plurality of one or more routers includes static route statements.

54. The method of claim 49 , wherein the altering of the plurality of one or more routing tables includes configuring a plurality of one or more routers, wherein the configuring the plurality of one or more routers adjust the processing of dynamic routing updates.

55. The method of claim 49 , wherein the altering of the plurality of one or more routing tables includes configuring a plurality of one or more routers, wherein the configuring the plurality of one or more routers adjust the processing of dynamic routing updates.

56. The method of claim 49 , wherein the altering of the plurality of one or more routing tables is performed at least partly by a plurality of one or more dynamic routing protocols.

57. The method of claim 56 , wherein the plurality of one or more dynamic routing protocols includes BGP.

58. The method of claim 56 , wherein the plurality of one or more dynamic routing protocols includes a plurality of one or more IGP routing protocols.

59. The method of claim 58 , wherein the plurality of one or more IGP routing protocols includes OSPF.

60. The method of claim 49 , wherein the altering of the plurality of one or more routing tables are applied manually by a user.

61. The method of claim 1 , wherein the rearchitecting of the internetwork includes redirecting the subset of flows to a second internetwork coupled to the internetwork at one or more Points of Presence.

62. The method of claim 61 , wherein the redirecting of the subset of flows is at least partly across an exit point traversed by at least a portion of the subset of flows from the internetwork, wherein the at least one performance metric is optimized.

63. The method of claim 1 , wherein the optimizing of the performance metric includes: calculating at least one performance metric for a plurality of one of more paths in the at least the portion of the internetwork; and at least partly responsive to the calculating of the at least one performance metric, selecting a path in the at least the portion of the internetwork from the plurality of one or more paths.

64. The method of claim 63 , wherein the selected path is a direct path.

65. The method of claim 63 , wherein the selected path is an indirect path.

66. The method of claim 1 , wherein the optimizing of the performance metric includes: selecting a plurality of one or more exit points; and selecting a plurality of one or more paths to reach the plurality of one or more exit points in the at least the portion of the internetwork.

67. The method of claim 66 , wherein the steps of selecting the plurality of one or more exit points and selecting the plurality of one or more paths are performed separately.

68. The method of claim 66 , wherein the steps of selecting the plurality of one or more exit points and selecting the plurality of one or more paths are performed together.

69. The method of claim 1 , wherein the measuring performance characteristics includes monitoring a first plurality of one or more flows in the internetwork, duplicating one or more packets from the plurality of one or more flows, and measuring the one or more duplicated packets.

70. The method of claim 1 , wherein the measuring performance characteristics includes serving as a proxy hop for a second plurality of one or more flows, and measuring the second plurality of one or more flows.

71. The method of claim 1 , wherein the measuring performance characteristics includes encoding material within a third plurality of one or more flows, causing a fourth plurality of one or more flows to be generated, wherein the third plurality of one or more flows traverses a first path of the internetwork, and the fourth plurality of one or more flows traverses a second path of the internetwork, wherein at least a portion of the first path of the internetwork and at least a portion of the second path of the internetwork are equal, and measuring a subset of the fourth plurality of one or more flows.

72. The method of claim 1 , wherein the measuring performance characteristics is done at least partly using flow information export.

73. The method of claim 1 , wherein the measuring performance characteristics is done using RMON II.

74. The method of claim 1 , wherein the measuring performance characteristics is provided by a source external to the subset of flows.

75. A method of optimizing traffic in an internetwork, the method comprising: selecting a subset of flows in the internetwork for monitoring, wherein the subset of flows includes one of one flow, some flows, and all flows; measuring performance characteristics of the subset of flows in at least a portion of the internetwork, the performance characteristics.including one or more of: a plurality of one or more round trip measurements for each of the subset of flows and a plurality of one or more one-way measurements for each of the subsets of flows, the measuring performance characteristics including: measuring a second plurality of one or more packets in the internetwork, wherein the second plurality of one or more packets were already launched; monitoring a first plurality of one or more flows in the internetwork, duplicating one or more packets from the plurality of one or more flows, and measuring the one or more duplicated packets; serving as a proxy hop for a second plurality of one or more flows, and measuring the second plurality of one or more flows, and encoding material within a third plurality of one or more flows, causing a fourth plurality of one or more flows to be generated, wherein the third plurality of one or more flows traverses a first path of the internetwork, and the fourth plurality of one or more flows traverses a second path of the internetwork, wherein at least a portion of the first path of the internetwork and at least a portion of the second path of the internetwork are equal, and measuring a subset of the fourth plurality of one or more flows; calculating at least one performance metric for the subset of flows in the at least a portion of the internetwork, the at least one performance metric at least partly determined from the measured performance characteristics; and in response to calculating the at least one performance metric, affecting the routing of the subset of flows by altering a plurality of one or more DNS entries in the internetwork.

76. A network systems, comprising: a plurality of one or more network devices configured such that, when the plurality of one or more network devices is deployed in an internetwork, the plurality of one or more network devices performs: selecting a subset of flows in the networks for monitoring, wherein the subset of flows includes one of one flow, some flows, and all flows; measuring performance characteristics of the subset of flows in at least a portion of the internetwork, the performance characteristics including one or more of: a plurality of one or more round trip measurements for each of the subset of flows and a plurality of one or more one-way measurements for each of the subset of flows, the measuring performance characteristics including: launching a first plurality of one or more packets, and measuring the first plurality of one or more packets; measuring a second plurality of one or more packets in the internetwork, wherein the second plurality of one or more packets were already launched; monitoring a first plurality of one or more flows in the internetwork, duplicating one or more packets from the plurality of one or more flows, and measuring the one or more duplicated packets; serving as a proxy hop for a second plurality of one or more flows, and measuring the second plurality of one or more flows; and encoding material within a third plurality of one or more flows, causing a fourth plurality of one or more flows to be generated, wherein the third plurality of one or more flows traverses a first path of the internetwork, and the fourth plurality of one or more flows traverses a second path of the internetwork, wherein at least a portion of the first path of the internetwork and at least a portion of the second path of the internetwork are equal, and measuring a subset of the fourth plurality of one or more flows; calculating at least one performance metric for the subset of flows in the at least a portion of the internetwork, the at least one performance metric at least partly determined from the measured performance characteristics; and in response to calculating the at least one performance metric, rearchitecting the internetwork to optimize one or more of the at least one performance metric using peer-to-peer communications, rearchitecting the internetwork including at least one of: altering a plurality of one or more routing tables in the internetwork, wherein the plurality of one or more routing tables include at least one of: network-layer routing tables, layer 3 routing tables, IP routing tables, layer 2 forwarding tables, and MPLS forwarding tables; redirecting the subset of flows to a second internetwork coupled to the internetwork at one or more Points of Presence; and affecting forwarding decisions of the subset of flows, by imposing one or more of: NAT, GRE, and tunneling techniques other than GRE.

77. The network system of claim 76 , where the rearchitecting of the internetwork is performed within the device.

78. A network systems, comprising: a plurality of one or more network devices configured such that, when the plurality of one more network devices is deployed in an internetwork, the plurality of one or more network devices performs: selecting a subset of flows in the internetwork for monitoring, wherein the subset of flows includes one or one flow, some flows, and all flows; measuring performance characteristics of the subset of flows in at least a portion of the internetwork, the performance characteristics including one or more of: a plurality of one or more round trip measurements for each of the subset of flows and a plurality of one or more one-way measurements for each of the subset of flows, the measuring performance characteristics including: measuring a second plurality of one or more packets in the internetwork, wherein the second plurality of one or more packets were already launched; monitoring a first plurality of one or more flows in the internetwork, duplicating one or more packets from the plurality of one or more flows, and measuring the one or more duplicated packets; serving as proxy hop for a the second plurality of one or more flows; and measuring the second plurality of one or more flows, and encoding material within a third plurality of one or more flows, causing a fourth plurality of one or more flows to be generated, wherein the third plurality of one or more flows traverses a first path of the internetwork, and the fourth plurality of one or more flows traverses a second path, of the internetwork, wherein at least a portion of the second path of the internetwork and at least a portion of the second path of the internetwork are equal, and measuring a subset of the fourth plurality of one or more flows; calculating at least one performance metric for the subset of flows in the at least a portion of the internetwork, the at least one performance metric at least partly determined from the measured performance characteristics; and in response to calculating the at least one performance metric, affecting the routing of the subset of flows by altering a plurality of one or more DNS entries in the internetwork.

79. The network system of claim 78 , where the affecting the routing of the subset of flows is performed within the device.

80. The method of claim 1 , wherein the subset of flows is selected based on a rate of incoming requests.

81. The method of claim 1 , wherein te subset of flows is selected based on applying thresholds to statistics related to the flows.

82. The method of claim 1 , wherein the subset of flows is based on costs of paths related to the flows, bandwidth consumed over a period of time, or both.

370/237

4284852 August 1981 Szybicki et al.
4345116 August 1982 Ash et al.
4495570 January 1985 Kitajima et al.
4594704 June 1986 Ollivier
4669113 May 1987 Ash et al.
4704724 November 1987 Krishnan et al.
4726017 February 1988 Krum et al.
4748658 May 1988 Gopal et al.
4788721 November 1988 Krishnan et al.
4839798 June 1989 Eguchi et al.
4920432 April 1990 Eggers et al.
4931941 June 1990 Krishnan
4939726 July 1990 Flammer et al.
4949187 August 1990 Cohen
4949248 August 1990 Caro
5142570 August 1992 Chaudhary et al.
5172413 December 1992 Bradley et al.
5253341 October 1993 Rozmanith et al.
5287537 February 1994 Newmark et al.
5291554 March 1994 Morales
5341477 August 1994 Pitkin et al.
5343463 August 1994 van Tetering et al.
5371532 December 1994 Gelman et al.
5375070 December 1994 Hershey et al.
5406502 April 1995 Haramaty et al.
5410343 April 1995 Coddington et al.
5414455 May 1995 Hooper et al.
5442389 August 1995 Blahut et al.
5442390 August 1995 Hooper et al.
5442749 August 1995 Northcutt et al.
5452294 September 1995 Natarajan
5467345 November 1995 Cutler, Jr. et al.
5471622 November 1995 Eadline
5471623 November 1995 Napolitano, Jr.
5475615 December 1995 Lin
5477536 December 1995 Picard
5508732 April 1996 Bottomley et al.
5515511 May 1996 Nguyen et al.
5519435 May 1996 Anderson
5521591 May 1996 Arora et al.
5528281 June 1996 Grady et al.
5535195 July 1996 Lee
5537394 July 1996 Abe et al.
5563875 October 1996 Hefel et al.
5574938 November 1996 Bartow et al.
5590126 December 1996 Mishra et al.
5629930 May 1997 Beshai et al.
5631897 May 1997 Pacheco et al.
5636216 June 1997 Fox et al.
5654958 August 1997 Natarajan
5659796 August 1997 Thorson et al.
5668800 September 1997 Stevenson
5675741 October 1997 Aggarwal et al.
5729528 March 1998 Salingre et al.
5754639 May 1998 Flockhart et al.
5787253 July 1998 McCreery et al.
5793976 August 1998 Chen et al.
5802106 September 1998 Packer
5805594 September 1998 Kotchey et al.
5812528 September 1998 VanDervort
5835710 November 1998 Nagami et al.
5845091 December 1998 Dunne et al.
5884047 March 1999 Aikawa et al.
5935216 August 1999 Benner et al.
5940478 August 1999 Vaudreuil et al.
5944779 August 1999 Blum
5974457 October 1999 Waclawsky et al.
6006264 December 1999 Colby et al.
6009081 December 1999 Wheeler et al.
6026441 February 2000 Ronen
6034946 March 2000 Roginsky et al.
6069889 May 2000 Feldman et al.
6078963 June 2000 Civanlar et al.
6108703 August 2000 Leighton et al.
6111881 August 2000 Soncodi
6119235 September 2000 Vaid et al.
6130890 October 2000 Leinwand et al.
6167052 December 2000 McNeill et al.
6173324 January 2001 D'Souza
6178448 January 2001 Gray et al.
6185598 February 2001 Farber et al.
6185601 February 2001 Wolff
6189044 February 2001 Thomson et al.
6226266 May 2001 Galand et al.
6275470 August 2001 Ricciulli
6282562 August 2001 Sidi et al.
6286045 September 2001 Griffths et al.
6292832 September 2001 Shah et al.
6317778 November 2001 Dias et al.
6339595 January 2002 Rekhter et al.
6341309 January 2002 Vaid et al.
6385643 May 2002 Jacobs et al.
6415323 July 2002 McCanne et al.
6434606 August 2002 Borella et al.
6438592 August 2002 Killian
6452950 September 2002 Ohlsson et al.
6453356 September 2002 Sheard et al.
6463454 October 2002 Lumelsky et al.
6493353 December 2002 Kelly et al.
6522627 February 2003 Mauger
6526056 February 2003 Rekhter
6556582 April 2003 Redi
6601098 July 2003 Case et al.
6601101 July 2003 Lee et al.
6608841 August 2003 Koodli
6611872 August 2003 McCanne
6614789 September 2003 Yazdani et al.
6625648 September 2003 Schwaller et al.
6631419 October 2003 Greene
6633640 October 2003 Cohen et al.
6661797 December 2003 Goel et al.
6687229 February 2004 Kataria et al.
6704768 March 2004 Zombek et al.
6707824 March 2004 Achilles et al.
6711137 March 2004 Klassen et al.
6711152 March 2004 Kalmanek, Jr. et al.
6714549 March 2004 Phaltankar
6728484 April 2004 Ghani
6748426 June 2004 Shaffer et al.
6751562 June 2004 Blackett et al.
6757255 June 2004 Aoki et al.
6760775 July 2004 Anerousis et al.
6760777 July 2004 Agarwal et al.
6766381 July 2004 Barker et al.
6795399 September 2004 Benmohamed et al.
6795860 September 2004 Shah
6801502 October 2004 Rexford et al.
6819662 November 2004 Grover et al.
6820133 November 2004 Grove et al.
6826613 November 2004 Wang et al.
6829221 December 2004 Winckles et al.
6836463 December 2004 Garcia-Luna-Aceves et al.
6839745 January 2005 Dingari et al.
6839751 January 2005 Dietz et al.
6885641 April 2005 Chan et al.
6909700 June 2005 Benmohamed et al.
6956858 October 2005 Hariguchi et al.
6963575 November 2005 Sistanizadeh et al.
6973490 December 2005 Robertson et al.
6981055 December 2005 Ahuja et al.
6993584 January 2006 Border et al.
7002917 February 2006 Saleh
7020086 March 2006 Juttner et al.
7043541 May 2006 Bechtolsheim et al.
7043562 May 2006 Dally et al.
7065584 June 2006 Shavitt et al.
7085230 August 2006 Hardy
7111073 September 2006 Jain et al.
7123620 October 2006 Ma
7155436 December 2006 Hegde et al.
7269157 September 2007 Klinker et al.
2001/0010059 July 2001 Burman et al.
2001/0026537 October 2001 Massey
2001/0037311 November 2001 McCoy et al.
2002/0038331 March 2002 Flavin
2002/0087687 July 2002 Zaifman et al.
2002/0101821 August 2002 Feldmann et al.
2002/0124100 September 2002 Adams
2002/0152318 October 2002 Menon et al.
2002/0184527 December 2002 Chun et al.
2003/0016770 January 2003 Trans et al.
2004/0030776 February 2004 Cantrell et al.
2005/0044270 February 2005 Grove et al.
2005/0083912 April 2005 Afshar et al.
2005/0132060 June 2005 Mo et al.
2005/0185654 August 2005 Zadikian et al.
2005/0201302 September 2005 Gaddis et al.
2005/0243726 November 2005 Narendran
2006/0026682 February 2006 Zakas
0 504 537 March 1991
0 528 075 February 1993
0 788 267 August 1997
0 598 969 February 1999
0 942 560 September 1999
0 982 901 January 2000
0 977 456 February 2000
0 999 674 April 2004
WO 94/08415 April 1994
WO 99/06913 February 1999
WO 99/14907 March 1999
WO 99/14931 March 1999
WO 99/14932 March 1999
WO 99/18751 April 1999
WO 99/30460 June 1999
WO 99/39481 August 1999
WO 00/04458 January 2000
WO 00/25224 May 2000
WO 00/38381 June 2000
WO 00/45560 August 2000
WO 00/52906 September 2000
WO 00/62489 October 2000
WO 00/72528 November 2000
WO 00/79362 December 2000
WO 00/79730 December 2000
WO 01/06717 January 2001
WO 01/13585 February 2001
WO 02/33896 April 2002

Z. Wang et al., “Resource Allocation for Elastic Traffic: Architecture and Mechanisms,” Conference Proceedings Article, 2000, XP010376681, pp. 159-169. cited by other
J. Yu, “Scalable Routing Design Principles,” Ref. No. RFC 2791, Network Working Group, Jul. 31, 2000, pp. 1-26. cited by other
Paul Francis et al., “An Architecture for a Global Internet Host Distance Estimation Service,” pp. 1-17. cited by other
T. Bates et al., “Multiprotocol Extensions for BGP-4”, XP-00219077, Jun. 2000, pp. 1-10. cited by other
S. Kumar et al., “The MASC/BGMP Architecture for Inter-domain Multicast Routing,” 12 pages. cited by other
S. Berson et al., “An Architecture for Advance Reservations in the Internet,” USC Information Sciences Institute, Jul. 16, 1998, pp. 1-21. cited by other
R. P. Draves et al., “Constructing Optimal IP Routing Tables,” 1999 IEEE, 1-10. cited by other
R. Govindan et al., “An Analysis of Internet Inter-Domain Topology and Route Stability,” USC Information Sciences Institute, 1997 IEEE, 8 pages. cited by other
V. Paxson, “Toward a Framework for Defining Internet Performance Metrics,” http://www.isoc.org/inet96/proceedings/d3/d3—3.htm, pp. 1-20. cited by other
C. Alaettinoglu et al. “Routing Policy Specification Language (RPSL),” http://quimby.gnus.org/internet-drafts/draft-ietf-rps-rpsl-v2-00.txt, pp. 1-56. cited by other
P. Traina, “BGP-4 Protocol Analysis,” Mar. 1995, pp. 1-10. cited by other
B. Krishnamurthy et al., “On Network-Aware Clustering of Web Clients,” 14 pages. cited by other
Sami Iren et al., “The Transport Layer: Tutorial and Survey”, XP-002210446, ACM Computing Surveys, vol. 31, No. 4, Dec. 1999, pp. 360-405. cited by other
D. B. Ingham et al., “Supporting Highly Manageable Web Services”, Computer Networks and ISDN Systems 29 (1997), pp. 1405-1416. cited by other
Nick Feamster et al., “Controlling the Impact of BGP Policy Changes on IP Traffic”, Nov. 6, 2001, pp. 1-18. cited by other
Dean Jones, “Developing Shared Ontologies in Multi-Agent Systems”, pp. 1-10. cited by other
L. Massoulié et al., “Arguments in favour of admission control for TCP flows”, pp. 1-16. cited by other
Yasushi Saito et al., “Manageability, Availability and Performance in Porcupine: A Highly Scalable, Cluster-Based Mail Service”, 17th ACM Symposium on Operating System Principles, pp. 1-15, Dec. 1999. cited by other
Nikolaos Skarmeas et al., “Content Based Routing as the Basis for Intra-Agent Communication”, Department of Computing Imperial College London. cited by other
Nikolaos Skarmeas et al., “Intelligent Routing Based on Active Patterns as the Basis for the Integration of Distributed Information Systems”, Department of Computing Imperial College London, Feb. 1997. cited by other
C. Alaettinoglu et al. “Routing Policy Specification Language (RPSL),” http://quimby.gnus.org/internet-drafts/draft-ietf-rps-rpsl-v2-00.txt, pp. 1-56, 1999. cited by other
B. Krishnamurthy et al., “On Network-Aware Clustering of Web Clients,” 14 pages, 2000. cited by other
Dean Jones, “Developing Shared Ontologies in Multi-Agent Systems”, pp. 1-10, 2001 cited by other
L. Massoulié et al., “Arguments in favour of admission control for TCP flows”, pp. 1-16, 2001. cited by other

edspgr.07336613