Congestion management and latency prediction in CSMA media

• US 8,943,219 B2
• Filed: 06/03/2013
• Issued: 01/27/2015
• Est. Priority Date: 02/01/2006
• Status: Active Grant

• Current Assignee: CoCo Communications, Corp.
• Original Assignee: CoCo Communications, Corp.
• Inventors: Eller, Riley, Edwards, Dennis, Bruestle, Jeremy, Tucker, Mark L
• Primary Examiner(s): SHINGLES, KRISTIE D
• Application Number:
  US13/908,924
• Publication Number:
  US 20130268659A1
• Time in Patent Office: 603 Days
• Field of Search: 709/224, 709/223, 709/235
• US Class Current: 709/235
• CPC Class Codes: H04L 12/413 with random access, e.g. carrier-sense multiple-access with collision detection [CSMA-CD]
• CPC Class Codes: H04L 41/14 Network analysis or design
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/11 Identifying congestion

Abstract

A facility for congestion management and latency prediction is described. In various embodiments, the facility sums a series of fractional transmission delays wherein each fractional transmission delay is measured as a probability of a failed transmission attempt multiplied by the cost of the failed transmission attempt, and provides the sum.

Claims

1. A method to determine data-transmission congestion, the method comprising:

generating an ordered series of bits, the ordered series of bits representing a history of communication receipts and communication failures between two network devices;

splitting the ordered series of bits into a first portion of ordered series of bits and a second portion of ordered series of bits, wherein the first portion of ordered series of bits corresponds to earlier communication receipts and communication failures, and wherein the second portion of ordered series of bits corresponds to later communication receipts and communication failures that occur after the first portion of ordered series of bits;

determining a first number that corresponds to a number of communication receipts and a number of communication failures during the first portion of ordered series of bits;

determining a second number that corresponds to a number of communication receipts and a number of communication failures during the second portion of ordered series of bits;

defining a null hypothesis that the difference between the first number and the second number is not statistically significant;

determining an amount of data-transmission congestion using the entire ordered series of bits when the null hypothesis is true; and

determining the amount of data-transmission congestion using the second portion of ordered series of bits when the null hypothesis is false.

2. A non-transitory computer-readable medium comprising executable program code that when executed by a processing element implements a method to determine data-transmission congestion between a first network node and a second network node, the method comprising:

monitoring a series of received and dropped messages communicated between the first network node and the second network node, wherein the received messages are associated with a first bit and the dropped messages are associated with a second bit;

generating a log of data as each new message is monitored, wherein the log of data is an ordered series of the first bits and the second bits ordered from a current bit associated with a current message to an oldest bit associated with an oldest message, wherein a first new bit or a second new bit associated with the new monitored message is added to the log of data as the current bit, and wherein the oldest bit is removed from the log of data;

dividing, as each new message is monitored, the log of data into a first log portion containing bits associated with messages monitored during a first time period and a second log portion containing bits associated with messages monitored during a second time period, wherein the second time period is earlier than the first time period;

determining, as each new message is monitored, a first number, wherein the first number corresponds to a number of the first bits in the first log portion;

determining, as each new message is monitored, a second number, wherein the second number corresponds to a number of the first bits in the second log portion;

if the first number is at least equal to the second number by a predetermined criterion, continuing to generate the log of data as a next message is provided; and

if the first number is less than the second number by the predetermined criterion, discarding the second log portion and continuing to generate the log of data using a new number of bits as the next message is provided, wherein the new number of bits corresponds to the number of the bits in the first log portion.

3. The medium of claim 2, wherein monitoring further includes monitoring the next message provided by the second network node to the first network node, wherein one of the first bit and the second bit is generated based on whether the next message is received or dropped, and wherein the generated bit is added as the current bit of the log of data.

4. The medium of claim 2, wherein the first and second log portions comprise two halves of the log of data.

5. The medium of claim 2, wherein each first bit is one of a first binary number that indicates that the message was received and each second bit is a second binary number that indicates that the message was dropped.

6. The medium of claim 2, wherein the first number corresponds to a ratio of first bits to a total number of bits in the first log portion, and wherein the second number corresponds to a ratio of first bits to a total number of bits in the second log portion.

7. A system, comprising:

a first network node configured to provide messages to a second network node; and

a third network node configured to;

monitor a series of received and dropped messages between the first network node and the second network node;

generate, as each new message is monitored, a log of data indicating whether each said message was received or dropped,

wherein the log of data comprises a total number of bits in an ordered series that is ordered from a current bit associated with a current message to an oldest bit associated with an oldest message,

wherein each bit is one of a first binary number indicating that the message was received and a second binary number indicating that the message was dropped,

wherein the bit associated with the new monitored message is added as the current bit into the log of data, and

wherein the oldest bit is removed from the log of data;

divide the log of data into a first log portion containing data associated with messages monitored during a first time period and a second log portion containing data associated with messages monitored during a second time period occurring earlier than the first time period;

determine, as each new message is monitored, a first number, wherein the first number corresponds to a number of first bits in the first log portion, or wherein the first number corresponds to a number of second bits in the first log portion;

determine, as each new message is monitored, a second number, wherein the second number corresponds to a number of the first bits in the second log portion when the first number corresponds to the number of the first bits in the first log portion, or wherein the second number corresponds to a number of the second bits in the second log portion when the first number corresponds to the number of the second bits in the first log portion;

if the first number is at least equal to the second number by a predetermined criterion, continuing to generate the log of data using the total number of bits as a next message is provided; and

if the first number is less than the second number by the predetermined criterion, discarding the second log portion and continuing to generate the log of data using a new number of bits as the next message is provided, wherein the new number of bits corresponds to a number of the bits in the first log portion.

8. The system of claim 7, wherein the first log portion and the second log portion are two halves of the log of data.

9. The system of claim 7, wherein the second log portion is discarded.

10. A method to determine data-transmission congestion between a first network node and a second network node, the method comprising:

monitoring a series of new messages between the first network node and the second network node, wherein received messages are associated with a first bit and dropped messages are associated with a second bit;

generating, in response to monitoring each new message, a log of data comprising a total number of bits corresponding to a predefined number of monitored messages, wherein the log of data is an ordered series of the first and second bits ordered from a current bit associated with a current message to an oldest bit associated with an oldest message, wherein a new bit associated with the new monitored message is added as the current bit into the log of data, and wherein a bit associated with the oldest message is removed from the log of data;

dividing, in response to monitoring each new message, the log of data into a first log portion containing a first number of bits associated with a first number of messages monitored during a first time period and a second log portion containing a second number of bits associated with a second number of messages monitored during a second time period, wherein the second time period is earlier than the first time period, and wherein the first number of bits associated with the first number of messages and the second number of bits associated with the second number of messages equals the total number of bits;

determining, in response to monitoring each new message, a first number corresponding to a number of the first bits in the first log portion;

determining, as each new message is monitored, a second number corresponding to a number of the first bits in the second log portion;

comparing, as each new message is monitored, the first number and the second number;

if the first number is at least equal to the second number by a predetermined criterion, continuing to generate the log of data using the total number of bits as a next message is monitored; and

if the first number is less than the second number by the predetermined criterion, discarding the second log portion and continuing to generate the log of data using a new number of bits as the next message is monitored, wherein the new number of bits corresponds to the first number of bits in the first log portion.

11. The method of claim 10, wherein each bit is one of a first binary number and a second binary, wherein the first binary number indicates that the associated message was received, wherein the second binary number indicates that the associated message was dropped.

12. The method of claim 10, wherein monitoring further includes monitoring the next message between the second network node to the first network node, wherein one of the first bit and the second bit is generated based on whether the next message is received or dropped, and wherein the generated bit is added as the current bit of the log of data.

13. A method to determine data-transmission congestion between a first network node and a second network node, wherein received messages are associated with a first bit and dropped messages are associated with a second bit, the method comprising:

generating, as each new message is monitored, a log of data comprising a total number of bits corresponding to a number of monitored messages, wherein the log of data is an ordered series of the first bits and the second bits ordered from a current bit associated with a current message to an oldest bit associated with an oldest message, wherein the bit associated with the new monitored message is added as the current bit into the log of data, and wherein a bit associated with the oldest message is removed from the log of data;

dividing, as each new message is monitored, the log of data into a first log portion containing the first bits and the second bits associated with messages monitored during a first time period and a second log portion containing the first bits and the second bits associated with messages monitored during a second time period, wherein the second time period is earlier than the first time period;

determining, as each new message is monitored, a first number corresponding to a number of the first bits in the first log portion;

determining, as each new message is monitored, a second number corresponding to a number of the first bits in the second log portion;

comparing, as each new message is monitored, the first number and the second number;

if the first number is at least equal to the second number by a predetermined criterion, continuing to generate the log of data using the total number of bits; and

if the first number is less than the second number by the predetermined criterion, discarding the second log portion and continuing to generate the log of data using a new number of bits as a next message is provided, wherein the new number of bits corresponds to the number of the first bits and the second bits in the first log portion.

14. The method of claim 13, wherein further comprising:

monitoring the next message between the second network node to the first network node, wherein one of the first bit and the second bit is generated based on whether the next message is received or dropped, and wherein the generated bit is added as the current bit of the log of data.

15. The method of claim 13, wherein the first number corresponds to a ratio of first bits to a total number of bits in the first log portion, and wherein the second number corresponds to a ratio of first bits to a total number of bits in the second log portion.

16. The method of claim 13, wherein each first bit is one of a first binary number that indicates that the message was received and each second bit is a second binary number that indicates that the message was dropped.