Claims
- 1. In a data processing network, a method of data flow control between nodes in the network comprising the steps of:
- determining a share of network capacity that a data source should have according to a critically damped second-order system; and
- determining a source node data rate using said share when a network congestion signal indicates no congestion, said source node data rate representing a data sending rate of a source node.
- 2. The method of claim 1 wherein said source node data rate r.sub.i (t+.DELTA.t) is calculated from:
- r.sub.i (t+.DELTA.t)=max{0, 2r.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 r.sub.i (t-.DELTA.t)-2k.sub.r [r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k, is a damping constant, and wherein said network congestion signal is provided from a network device to said source node.
- 3. The method of claim 2 wherein said share S(t+.DELTA.t) is calculated from:
- S(t-.DELTA.t)=max{0, 2S)(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)+f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.s is a damping constant, and f.sub.1 is a preselected constant.
- 4. The method of claim 3 wherein said damping rate, k.sub.r, is in the range from 0 to 1, and said damping rate, k.sub.s is in the range from 0 to 1.
- 5. The method of claim 3 wherein said f.sub.1 is equal to a maximum link rate in the network.
- 6. The method of claim 1 wherein said source node data rate r.sub.i (t+.DELTA.t) is calculated from:
- r.sub.i (t-.DELTA.t)=max{0, 2.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 [r.sub.i (t-.DELTA.t)-S(t)]-2k.sub.r [r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.r is a damping constant, and said network congestion signal is provided from a network device to said source node.
- 7. The method of claim 5 wherein said share S(t+.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.2 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.s is a damping constant, and f.sub.2 is a preselected target share value.
- 8. The method of claim 7 wherein said damping rate, k.sub.s, is in the range from 0 to 1, and said damping rate, k.sub.s is in the range from 0 to 1.
- 9. The method of claim 7 wherein said f.sub.2 is equal to a maximum link rate in the network.
- 10. The method of claim 1 wherein said share S(t+.DELTA.t) is calculated from:
- S(t-.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.s is a damping constant, f.sub.1 is a preselected constant, and said network congestion signal is provided from a network device to said source node.
- 11. The method of claim 1 wherein said share S(t-.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.2 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.r is a damping constant, and f.sub.2 is a preselected constant representing a target share value, and wherein said network congestion signal is provided from a network device to said source node.
- 12. In a data processing network, a data processing system for data flow control between nodes in the network comprising:
- circuitry for determining a share of network capacity that a data source should have according to a critically damped second-order system; and
- circuitry coupled and responsive to said share determine circuitry for determining a source node data rate using said share when a network congestion signal indicates no congestion, said source node data rate representing a data sending rate of a source node.
- 13. The data processing system of claim 12 wherein said source data rate r.sub.i (t-.DELTA.t) is calculated from:
- r.sub.i (t-.DELTA.t)=max{0, 2r.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 r.sub.i (t-.DELTA.t)-2k.sub.r [r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.r is a damping constant, and said network congestion signal is provided from a network device to said source node.
- 14. The data processing system of claim 13 wherein said share S(t+.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.s is a damping constant and f.sub.1 is a preselected constant.
- 15. The data processing system of claim 14 wherein said f.sub.1 is equal to a maximum network link rate.
- 16. The data processing system of claim 12 wherein said source node data rate is r.sub.i (t+.DELTA.t) calculated from:
- r.sub.i (t-.DELTA.t)=max{0, 2r.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 [r.sub.i (t-.DELTA.t)-S(t)]-2k.sub.r [r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.r is a damping constant, and said network congestion signal is provided from a network device to said source node.
- 17. The data processing system of claim 16 wherein said share S(t+.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-s(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.2 ]-2k.sub.s [S(t)-s(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.s is a damping constant, and f.sub.2 is a preselected target share value.
- 18. The data processing system of claim 17 wherein said f.sub.2 is equal to a maximum network link rate.
- 19. The data processing system of claim 12 wherein said share S(t+.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)+f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.2 is a damping constant and f.sub.1 is a preselected constant, and said network congestion signal is provided from a network device to said source node.
- 20. The data processing system of claim 12 wherein said share S(t+.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.2 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.s is a damping constant, f.sub.2 is a preselected target share value, and said network congestion signal is provided from a network device to said source node.
- 21. A program product adaptable for storage on a storage media, said program product operable for determining data flow control from a source node to a destination node within a data processing network comprising:
- programming for determining a share of network capacity that a data source should have according to a critically damped second-order system; and
- programming for determining a source node data rate using said share when a network congestion signal indicates no congestion, said source node data rate representing a data sending rate of a source node.
- 22. The program product adaptable for storage on a storage media of claim 21 wherein said source node data rate is calculated from:
- r.sub.i (t+.DELTA.t)=max{0, 2r.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 r.sub.i (t-.DELTA.t)-2k.sub.r [r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.r is a damping constant, and said network congestion signal is provided from a network device to said source node.
- 23. The program product adaptable for storage on a storage media of claim 22 wherein said share S(t+.DELTA.t) is calculated from:
- S(t-.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)-f.sub.1 ]-2k.sub.s[S (t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.s is a damping constant, and f.sub.1 is a preselected constant.
- 24. The program product available for storage on a storage media of claim 23 wherein said f.sub.1 is equal to a maximum link rate in the network.
- 25. The program product adaptable for storage on a storage media of claim 21 wherein said source node data rate r.sub.i (t+.DELTA.t) is calculated from:
- r.sub.i (t+.DELTA.t)=max{0, 2r.sub.i (t)-r.sub.i (t)-r.sub.i (t-.DELTA.t)-k.sub.r.sup.2 [r.sub.i (t-.DELTA.t)-S(t)]-2k.sub.r ]r.sub.i (t)-r.sub.i (t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.r is a damping constant, and said network congestion signal is provided from a network device to said source node.
- 26. The program product adaptable for storage on a storage media of claim 25 wherein said share S(t-.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S)(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)+f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.s is a damping constant and f.sub.2 is a preselected target share value.
- 27. The program product adaptable for storage on a storage media of claim 26 wherein said f.sub.2 is equal to a maximum link rate in the network.
- 28. The program product adaptable for storage on a storage media of claim 21 wherein said share S(t-.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)+f.sub.1 ]-2k.sub.s [S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates network congestion, wherein k.sub.s is a damping constant, f.sub.1 is a preselected constant, and said network congestion signal is provided from a network device to said source node.
- 29. The program product adaptable for storage on a storage media of claim 21 wherein said share S(t-.DELTA.t) is calculated from:
- S(t+.DELTA.t)=max{0, 2S(t)-S(t-.DELTA.t)-k.sub.s.sup.2 [S(t-.DELTA.t)+f.sub.1 ]-2k.sub.s S(t)-S(t-.DELTA.t)]}
- when said network congestion signal indicates no network congestion, wherein k.sub.s is a damping constant, f.sub.2 is a preselected target share value, and said network congestion signal is provided from a network device to said source node.
CROSS-REFERENCE TO RELATED APPLICATIONS
The present invention is related to U.S. patent application Ser. No. 08/977,475 filed Nov. 24, 1997 and U.S. patent application Ser. No. 08/978,178 filed Nov. 25, 1997, which are hereby incorporated by reference herein.
US Referenced Citations (9)