Network latency analysis packet and method

• US 20070274227A1
• Filed: 05/24/2006
• Published: 11/29/2007
• Est. Priority Date: 05/24/2006
• Status: Active Grant

• Current Assignee: AT&T Corporation (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Werner, Carl E., Rauscher, Frank
• Granted Patent: US 7,961,635 B2
• Time in Patent Office: Days
• US Class Current: 370/252
• CPC Class Codes: H04L 41/0213 Standardised network management protocols, e.g. simple network management protocol [SNMP]
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 43/0858 One way delays
• CPC Class Codes: H04L 43/106 using time related information in packets, e.g. by adding timestamps
• CPC Class Codes: H04L 43/50 Testing arrangements

Abstract

A packet and a method for analyzing network latency are disclosed. The disclosed systems and methods measure the latency between nodes in a network and do so while using less bandwidth and processing than traditional methods by using a packet to traverse a network and collect timestamps at various nodes so that the delay in transit time between nodes can be calculated when the packet returns to a server.

Claims

1. A method of measuring network latency, the method comprising the steps of:

(a) creating a packet at a source server, the packet comprising data including a sequence of nodes for the packet to traverse;

(b) sending the packet to a first node in the sequence of nodes for sequential transmission to each remaining node in the sequence of nodes; and

(c) receiving the packet at the source server after the packet has traversed the nodes of the sequence.

2. The method of claim 1, the method further comprising the step of:

(d) specifying a class of service for the packet.

3. The method of claim 1, the method further comprising the step of:

(e) calculating a delay between at least one pair of nodes in the network based on timestamps entered in the packet by each of the nodes of the sequence.

4. The method of claim 3, wherein step (e) comprises subtracting a first timestamp associated with a first node in the sequence of nodes stored in the packet from a second timestamp associated with a second node following the first node in the sequence of nodes stored in the packet.

5. The method of claim 1, wherein (c) comprises marking the packet with a timestamp in a way such that the timestamp is associated with a node.

6. The method of claim 3, the method further comprising the step of:

(f) storing the delay calculated in (e) in a memory.

7. The method of claim 6, the method further comprising the step of:

(g) generating at least one point for a service level agreement graph based on timestamps entered in the packet by each of the nodes of the sequence.

8. The method of claim 7, the method further comprising the step of:

(h) generating the service level agreement graph.

9. The method of claim 1, wherein the packet is sent using internet protocol.

10. The method of claim 2, the method further comprising the step of:

(e) calculating a time difference between a first timestamp from a first node in the sequence of nodes and a last timestamp from a last node in the sequence of nodes by subtracting the first timestamp from the last timestamp.

11. A method for processing a packet in a network at a node, the method comprising the steps of:

(a) receiving the packet;

(b) parsing the packet to obtain data about the packet;

(c) reading a field in the packet to determine whether the packet is an administrative packet used for measuring network latency;

(d) marking the packet with a timestamp;

(e) determining a next node for the packet to be sent to; and

(f) sending the packet to the next node in the network.

12. The method of claim 11, wherein step (c) comprises examining a flag to determine whether the packet is an administrative packet used for measuring network latency.

13. The method of claim 11, wherein step (c) comprises examining a field in the packet to determine whether the packet is an administrative packet used for measuring network latency.

14. The method of claim 11, wherein step (b) comprises parsing the class of service information.

15. The method of claim 11, wherein the packet is marked with a timestamp when the packet is received by the node and the packet is marked with a second timestamp when the node is sending the packet to the next node in the network.

16. The method of claim 11, wherein step (e) comprises determining a second node in the network that is listed immediately after a first identifier of the node processing the packet.

17. The method of claim 11, wherein step (e) comprises:

(1) determining whether the packet has enough room to obtain an additional timestamp for a second node;

(2) determining a next node by randomly selecting the second node in the network that is connected to the node processing the packet if the packet has enough room to obtain an additional timestamp for a second node; and

(3) sending the packet back to a server that can process data from the packet to determine network delay if the packet does not have enough room to obtain an additional timestamp for the second node.

18. The method of claim 11, wherein the packet is received using internet protocol.

19. The method of claim 11, wherein step (e) comprises:

(1) determining whether the packet has a multiplier associated with the node;

(2) determining that a next node for the packet to be sent to is a node previously visited by the packet; and

(3) indicating that the packet has returned to the node previously visited by the packet by decrementing the multiplier associated with the node.

20. A method of measuring network latency, the method comprising the steps of:

(a) creating a packet with data including a sequence of nodes for the packet to traverse;

(b) sending the packet from a source to a first node in the sequence of nodes for sequential transmission to each remaining node in the sequence of nodes;

(c) receiving the packet at a node;

(d) determining whether the packet is an administrative packet used for measuring network latency;

(e) marking the packet with a first timestamp when the packet is received;

(f) determining a destination node for the packet by finding a next node identifier in the sequence of nodes;

(g) marking the packet with a second timestamp when the packet is leaving the node;

(h) sending the packet to the destination node;

(i) receiving the packet at a server when the packet has traversed the network via the sequence of nodes specified;

(j) parsing the data of the packet to obtain timestamps associated with nodes; and

(k) calculating a delay between nodes in the network based on the timestamps.