Flow control method using a dynamic major reduction factor

• US 6,400,684 B1
• Filed: 09/01/1998
• Issued: 06/04/2002
• Est. Priority Date: 09/01/1998
• Status: Expired due to Fees

• Current Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Benmohamed, Lotfi, Wang, Yung-Terng, Lau, Wing Cheong
• Primary Examiner(s): Nguyen, Chau
• Assistant Examiner(s): TRINH, DUNG N
• Application Number:
  US09/145,351
• Time in Patent Office: 1,372 Days
• Field of Search: 370/229, 370/230, 370/230.1, 370/232, 370/236.1, 370/395.1, 370/395.21, 370/395.43, 370/468, 370/233, 370/234
• US Class Current: 370/230.1
• CPC Class Codes: H04L 2012/5635 Backpressure, e.g. for ABR
• CPC Class Codes: H04L 2012/5682 Threshold; Watermark
• CPC Class Codes: H04Q 11/0478 Provisions for broadband connections

Abstract

A flow control method for available bit rate service in an asynchronous transfer mode network for detecting congestion in a switch and conveying congestion status information using resource management cells to the source which adjusts its flow rate accordingly. During detection of congestion in the switch the explicit rate field of the resource management cell is updated using a dynamic major reduction factor to improve transient queue-length performance.

Claims

1. A method for adjusting a flow rate in an asynchronous transfer mode network based on congestion information detected at a switch and conveyed using resource management cells having an explicit rate field, said method comprising the steps of:

determining upon an occurrence of an event an estimated fairshare of bandwidth allotted a particular connection at the switch, said determining step comprising the steps of;

detecting a maximum flow rate among all connections arriving at the switch upon arrival of a forward resource management cell at the switch; and

adjusting the detected maximum flow rate based on a running exponential weighted average; and

when a queue-length of the switch is greater than or equal to a queue-length threshold, adjusting the estimated fairshare of bandwidth allotted to the particular connection based on a dynamic major reduction factor, said dynamic major reduction factor being a non-increasing function of queue-length.

2. A method in accordance with claim 1, wherein said determining step comprises the step of defining the event based on arrival of a forward resource management cell at the switch.

3. A method in accordance with claim 1, wherein said determining step comprises the step of defining the event as a predetermined period of time.

4. A method for adjusting a flow rate in an asynchronous transfer mode network based on congestion information detected at a switch and conveyed using resource management cells having an explicit rate field, said method comprising the steps of:

determining upon an occurrence of an event an estimated fairshare of bandwidth allotted a particular connection at the switch; and

when a queue-length of the switch is greater than or equal to a queue-length threshold, adjusting the estimated fairshare of bandwidth allotted to the particular connection based on a dynamic major reduction factor, said dynamic major reduction factor being a non-increasing function of queue-length, said adjusting step comprising the step of calculating the explicit rate field as

5. A method for adjusting a flow rate in an asynchronous transfer mode network based on congestion information detected at a switch and conveyed using resource management cells having an explicit rate field, said method comprising the steps of:

determining upon an occurrence of an event an estimated fairshare of bandwidth allotted a particular connection at the switch; and

when a queue-length of the switch is greater than or equal to a queue-length threshold, adjusting the estimated fairshare of bandwidth allotted to the particular connection based on a dynamic major reduction factor, said dynamic major reduction factor being a non-increasing function of queue-length, said adjusting step comprising the step of defining the major reduction factor function to include a maximum major reduction factor value less than or equal to “

1” and

a minimum major reduction factor value greater or equal to “

0”

.

6. A method in accordance with claim 5, wherein said adjusting step comprises the step of defining the major reduction factor function so that the minimum major reduction factor value corresponds to a maximum queue-length value defined as

7. A method in accordance with claim 6, wherein said adjusting step comprises the step of defining the major reduction factor function as

MRF=MRF_max, for 0≦

q≦

q_lo;

MRF=ƒ

₁(q, q_lo, q_med, MRF_med, MRF_max, k₁), for q_lo<

q<

q_med;

MRF=MRF_med, for q=q_med;

MRF=ƒ

₂(q, q_med, q_hi, MRF_min, MRF_med, k₂), for q_med<

q<

q_hi;

MRF=MRF_min, for q≧

q_hi;

where

MRF_max is the maximum major reduction factor value;

MRF_min is the minimum major reduction factor value;

MRF_med is an intermediate major reduction factor value, where MRF_min<

MRF_med<

MRF_max;

q is the queue-length;

q_hi is the maximum queue-length;

q_lo is a minimum queue-length;

q_med is an intermediate queue-length, where q_lo<

q_med<

_hi;

ƒ

₁ and ƒ

₂ are first and second non-increasing functions of queue-length; and

k₁ and k₂ are constants.

8. A method in accordance with claim 7, wherein said adjusting step further comprises selecting the first and second non-increasing functions from the group consisting of an exponential concave function, an exponential convex function, a quadratic concave function, a quadratic convex function, and a linear function.

9. A method in accordance with claim 7, wherein the first and second non-increasing functions are the same function.

10. A method in accordance with claim 8, wherein said adjusting step comprises the step of defining at least one of the first and second non-increasing functions as the exponential convex function defined as

11. A method in accordance with claim 8, wherein said adjusting step comprises the step of defining at least one of the first and second non-increasing functions as the exponential concave function defined as

12. A method in accordance with claim 8, wherein adjusting step comprises defining at least one of the first and second non-increasing functions as the quadratic convex function defined as

13. A method in accordance with claim 8, wherein said adjusting step comprises the step of defining at least one of the first and second non-increasing functions as the quadratic concave function defined as

14. A method in accordance with claim 8, wherein said adjusting step comprises the step of defining at least one of the first and second non-increasing functions as the linear function defined as