Systems and methods for dropping data using a drop profile

• US 7,283,470 B1
• Filed: 07/30/2002
• Issued: 10/16/2007
• Est. Priority Date: 01/25/2002
• Status: Active Grant

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Basu, Debashis, Godbole, Avanindra, Sindhu, Pradeep, Boddu, Jayabharat
• Primary Examiner(s): Phunkulh; Bob A.
• Application Number:
  US10/207,002
• Time in Patent Office: 1,904 Days
• Field of Search: None
• US Class Current: 370/230
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/23 Bit dropping
• CPC Class Codes: H04L 47/30 in combination with information about buffer occupancy at either end or at transit nodes
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 47/326 with random discard, e.g. random early discard [RED]
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 49/9078 using an external memory or storage device

Abstract

A system selectively drops data from queues. The system includes a drop table that stores drop probabilities. The system selects one of the queues to examine and generates an index into the drop table to identify one of the drop probabilities for the examined queue. The system then determines whether to drop data from the examined queue based on the identified drop probability.

Claims

1. A system for selectively dropping data from a queue, comprising:

a plurality of queues configured to temporarily store data; and

a drop engine configured to select one of the queues to examine, the drop engine comprising;

a drop table configured to store a plurality of drop probabilities,

indexing logic configured to generate an index into the drop table to identify one of the drop probabilities, wherein the indexing logic is configured to;

determine a static amount of memory allocated to the examined queue,

determine an amount of memory used by the examined queue, and

determine the index into the drop table based on the static amount of memory allocated to the examined queue and the amount of memory used by the examined queue, and

drop decision logic configured to determine whether to drop data from a head of the examined queue based on the identified drop probability.

2. The system of claim 1, wherein the indexing logic is configured to:

generate the index into the drop table based on an amount of memory used by the examined queue, and

identify one of the drop probabilities in the drop table using the generated index.

3. The system of claim 1, wherein when determining the index, the indexing logic is configured to:

generate the index based on a ratio of the amount of memory used by the examined queue to a maximum of the static amount of memory allocated to the examined queue or the amount of memory that should be made available to the examined queue.

4. The system of claim 1, wherein the drop table includes:

a plurality of drop tables each of which is configured to store a plurality of drop probabilities.

5. The system of claim 4, wherein the indexing logic is further configured to:

select one of the drop tables, and

identify one of the drop probabilities in the selected drop table using the generated index.

6. The system of claim 5, wherein when selecting one of the drop tables, the indexing logic is configured to:

use one or more attributes of the data to select among the drop tables.

7. The system of claim 6, wherein the attributes include one or more of a Transmission Control Protocol or a Packet Level Protocol relating to the data.

8. The system of claim 1, wherein the drop decision logic includes:

a random number generator configured to generate a random number, and

a comparator configured to;

compare the identified drop probability to the random number, and

generate a drop decision based on a result of the comparison.

9. The system of claim 8, wherein the random number generator is a pseudo random number generator.

10. The system of claim 8, wherein the random number generator is a linear feedback shift register.

11. The system of claim 8, wherein the random number has a same number of bits as a number of bits in the identified drop probability.

12. The system of claim 8, wherein the random number has a greater number of bits than a number of bits in the identified drop probability.

13. The system of claim 8, wherein the drop decision logic further includes:

a logic operator configured to;

perform a logical operation on the result of the comparison and a signal based on an attribute extracted from the data, and

output, based on the logical operation, a decision signal that indicates whether to drop data from the examined queue.

14. The system of claim 13, wherein the logic operator is an AND gate.

15. The system of claim 1, wherein the drop decision logic is configured to:

determine whether to drop data from the examined queue based on the identified drop probability and other information including information extracted from the data.

16. The system of claim 15, wherein the information includes an attribute that indicates that the data should not be dropped.

17. The system of claim 1, wherein an index threshold is established;

and

wherein the drop decision logic is configured to;

determine not to drop the data when the index generated by the indexing logic is below the index threshold.

18. A system for selectively dropping data from a queue, comprising:

a plurality of queues configured to temporarily store data; and

a drop engine configured to select one of the queues to examine, the drop engine comprising;

a drop table configured to store a plurality of drop probabilities,

indexing logic configured to generate an index into the drop table to identify one of the drop probabilities, and

drop decision logic configured to determine whether to drop data from a head of the examined queue based on the identified drop probability, wherein the drop probability is based on a ratio of an amount of memory being used by the examined queue to a maximum of a static amount of memory allocated to the examined queue or an amount of memory that should be made available to the examined queue.

19. The system of claim 18, wherein the drop decision logic is configured to:

identify the data for dropping when the amount of memory being used by the examined queue is greater than the maximum of the static amount of memory allocated to the examined queue or the amount of memory that should be made available to the examined queue.

20. A system for selectively dropping data from a queue, comprising:

a plurality of queues configured to temporarily store data; and

a drop engine configured to select one of the queues to examine, the drop engine comprising;

a drop table configured to store a plurality of drop probabilities,

indexing logic configured to generate an index into the drop table to identify one of the drop probabilities, and

drop decision logic configured to determine whether to drop data from a head of the examined queue based on the identified drop probability, wherein the drop probability is based on a ratio of an amount of memory being used by the examined queue to a static amount of memory allocated to the examined queue.

21. A method for selectively dropping data from each of a plurality of queues that temporarily store data, comprising:

selecting one of the queues to examine;

providing a drop table that stores a plurality of drop probabilities;

generating an index into the drop table to identify one of the drop probabilities for the examined queue, wherein the generating the index includes;

determining a static amount of memory allocated to the examined queue,

determining an amount of memory used by the examined queue, and

determining the index into the drop table based on the static amount of memory allocated to the examined queue and the amount of memory used by the examined queue; and

determining whether to drop data from a head of the examined queue based on the identified drop probability.

22. The method of claim 21, wherein the generating an index includes:

determining an amount of memory used by the examined queue,

generating the index into the drop table based on the determined amount of memory, and

identifying one of the drop probabilities in the drop table using the generated index.

23. The method of claim 21, wherein the determining the index includes:

generating the index based on a ratio of the amount of memory used by the examined queue to a maximum of the static amount of memory allocated to the examined queue or the amount of memory that should be made available to the examined queue.

24. The method of claim 21, wherein the drop table includes a plurality of drop tables each of which is configured to store a plurality of drop probabilities;

and

wherein the method further comprises;

selecting one of the drop tables, and

identifying one of the drop probabilities in the selected drop table using the generated index.

25. The method of claim 24, wherein the selecting one of the drop tables includes:

using one or more attributes of the data to select among the drop tables.

26. The method of claim 25, wherein the attributes include one or more of a Transmission Control Protocol or a Packet Level Protocol relating to the data.

27. The method of claim 21, wherein the determining whether to drop data from the examined queue includes:

generating a random number,

comparing the identified drop probability to the random number, and generating a drop decision based on a result of the comparison.

28. The method of claim 27, wherein the random number has a same number of bits as the identified drop probability.

29. The method of claim 27, wherein the random number has a greater number of bits than the identified drop probability.

30. The method of claim 27, wherein the determining whether to drop data from the examined queue further includes:

performing a logical operation on the result of the comparison and a signal based on an attribute extracted from the data, and

outputting, based on the logical operation, a decision signal that indicates whether to drop data from the examined queue.

31. The method of claim 30, wherein the performing a logical operation includes:

ANDing the result of the comparison and the signal that indicates whether the identified drop probability is greater than zero.

32. The method of claim 21, wherein the determining whether to drop data includes:

determining whether to drop the data from the examined queue based on the identified drop probability and other information including information extracted from the data.

33. The method of claim 21, wherein the information includes an attribute that indicates that the data should not be dropped.

34. The method of claim 21, further comprising:

establishing an index threshold; and

wherein the determining whether to drop data includes;

determining not to drop the data when the index is below the index threshold.

35. A method for selectively dropping data from each of a plurality of queues that temporarily store data, comprising:

selecting one of the queues to examine;

providing a drop table that stores a plurality of drop probabilities;

generating an index into the drop table to identify one of the drop probabilities for the examined queue; and

determining whether to drop data from a head of the examined queue based on the identified drop probability, wherein the drop probability is based on a ratio of an amount of memory being used by the examined queue to a maximum of a static amount of memory allocated to the examined queue or an amount of memory that should be made available to the examined queue.

36. The method of claim 35, wherein the determining whether to drop data includes:

identifying the data for dropping when the amount of memory being used by the examined queue is greater than the maximum of the static amount of memory allocated to the examined queue or the amount of memory that should be made available to the examined queue.

37. A method for selectively dropping data from each of a plurality of queues that temporarily store data, comprising:

selecting one of the queues to examine;

providing a drop table that stores a plurality of drop probabilities;

generating an index into the drop table to identify one of the drop probabilities for the examined queue; and

determining whether to drop data from a head of the examined queue based on the identified drop probability, wherein the identified drop probability is based on a ratio of an amount of memory being used by the examined queue to a static amount of memory allocated to the examined queue.