Message network monitoring

• US 5,432,776 A
• Filed: 09/30/1993
• Issued: 07/11/1995
• Est. Priority Date: 10/01/1992
• Status: Expired due to Fees

• Current Assignee: Enterasys Networks Incorporated (Extreme Networks, Inc.)
• Original Assignee: Digital Equipment Corporation (HP Inc.)
• Inventors: Harper, John A.
• Primary Examiner(s): Chin, Wellington
• Application Number:
  US08/129,513
• Time in Patent Office: 649 Days
• Field of Search: 370/13, 370/17, 370/58.1, 370/58.2, 370/58.3, 370/60, 370/60.1, 370/94.1, 370/94.2, 370/110.1, 370/68.1, 379/112, 379/113, 379/133, 379/134, 379/137
• US Class Current: 370/252
• CPC Class Codes: G06F 2221/2101 Auditing as a secondary aspect
• CPC Class Codes: H04L 12/14 Charging , metering or billing arrangements for data wireline or wireless communications
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level

Abstract

A digital message routing network has at least some nodes at which messages are monitored for accounting purposes. Such a node contains means for generating and compressing accounting digests, each of which consists of a key K representing the nature and route of the message and a value V associated with the message. For each message, a digest is produced in register 30, and copied into a location in a first memory 22D at a location determined by hashing its key; further digests with the same key are compressed into the digest already in that location. A second memory 23D is operated as a cyclic buffer with head and tail counters HDCT 40 and TLCT 41. The address of the head of the buffer, ie the valid area of memory 23D, is included as an index I in each digest in the first memory 22D as that digest is created. Each time a digest is created or changed in the first memory 22D, it is copied into the second memory at the location in the second memory determined by its index. Digests are read out from the second memory in blocks; the corresponding digests in the first memory thereby become invalid, and have to be restarted by initializing their values and giving them new indexes.

Claims

1. In a link unit in a node in a digital message routing network, said node being one of a plurality of nodes, a digest processing means for generating and compressing a plurality of accounting digests, each of said digests having a unique key representing the nature and route of a corresponding message and a value associated with the message, characterized in that the digest processing means comprise:

a first memory into which said digests are written and compressed at locations determined by their corresponding unique keys;

a second memory operated as a cyclic buffer, an address of a head of said cyclic buffer included in each of said digests in the first memory, said head providing an index; and

copying means for each time one of said digests is created or changed in the first memory, copying it into the second memory at a location in the second memory determined by the index of the one of said digests.

2. The digest processing means according to claim 1, further comprising:

a validity test logic fed when the one of said digests is to be created or changed in the first memory, with the index of the digest and the address of the head and an address of a tail of the cyclic buffer to determine whether the corresponding location in the second memory is valid and, if not, to cause a new digest to be created in the first memory.

3. The digest processing means according to claim 2 further comprising a cycle counter, the contents of which are concatenated with an address of the head of the second memory to form an updated digest index.

4. The digest processing means according to claim 3 further comprising:

a microprocessor coupled to the first and second memories, said microprocessor used to handle a processing load of said link unit; and

a third, instruction, memory coupled to the microprocessor.

5. The digest processing means according to claim 4, characterized by a digest memory receiving the updated digest index from the second memory of the link unit and a management processor which compresses into the digest memory digests from all nodes.

6. The digest processing means according to claim 2 further comprising:

a microprocessor coupled to the first and second memories, the microprocessor used to handle the processing load of said link unit; and

a third, instruction, memory coupled to the microprocessor.

7. The digest processing means according to claim 6, characterized by a digest memory receiving the updated digest index from the second memory of the link unit and a management processor which compresses into the digest memory digests from all nodes.

8. The digest processing means according to claim 1 further comprising:

a microprocessor coupled to the first and second memories, the microprocessor used to handle the processing load of the link unit; and

a third, instruction, memory coupled to the microprocessor.

9. The digest processing means according to claim 8, characterized by a digest memory receiving the updated digest index from the second memory of the link unit and a management processor which compresses into the digest memory digests from all nodes.

10. A method of generating and compressing a plurality of accounting digests in a link unit in a node in a digital message routing network, said network having a plurality of nodes, each of said nodes having a plurality of link units, each one of said digests consisting of a unique key representing the nature and route of a corresponding message and a value associated with the corresponding message, comprising the steps of:

writing and compressing the digests at locations determined by their unique keys in a first memory;

as each one of said digests in the first memory is created, including in said digest an address of the head of a second memory operated as a cyclic buffer as an index; and

each time one of said digests is created or changed in the first memory, copying it a copying means into the second memory at a location in the second memory determined by the index of the digest.

11. The method of generating and compressing a plurailty of accounting digests according to claim 10 including the step of:

testing, when a new digest is to be created or changed in the first memory, its validity using an index of the digest and the address of the head and an address of a tail of the cyclic buffer to determine whether the corresponding location in the second memory is valid and, if not, to cause a new digest to be created in the first memory.

12. The method of generating and compressing a plurality of accounting digests according to claim 11 further including the step of:

concatenating the contents of a cycle counter with the address of the head of the second memory to form the index of the digest.

13. The method of generating and compressing a plurality of accounting digests according to claim 12 further including the step of:

compressing memory digests from the plurality of link units into a node digest memory fed from the second memories of the link units by a management processor.

14. The method of generating and compressing a plurality of accounting digests according to claim 11 further including the step of:

compressing memory digests from the plurality of link units into a node digest memory fed from the second memories of the link units by a management processor.

15. The method of generating and compressing a plurality of accounting digests according to claim 10 further including the step of:

compressing memory digests from the plurality of link units into a node digest memory fed from the second memories of the link units by a management processor.