Granted Patent B2

US 10,979,335 · App. 15/417,978 · Granted Apr 13, 2021

Inventors: Na Li (Austin, TX); Lan Li (Austin, TX)

Assignee: SOLARWINDS WORLDWIDE, LLC

H04L43/10H04L43/087H04L43/0829

Abstract

Claims (44)

1. A method comprising:

discovering, using a probe packet, multiple paths across a plurality of nodes in a network between a user equipment and an endpoint;

deriving, at the user equipment, performance metrics of an edge located between two of the plurality of nodes, wherein the performance metrics are derived based on performance of the two nodes bordering the edge, wherein the two nodes are located within one of the multiple paths;

determining a location of the edge where the performance metrics indicates a problem with the edge;

sending an indication to a network provider with the location of the edge, wherein the location is used to signal a problem with one of the two nodes bordering the edge;

calculating the performance metrics based on a first performance metric and a second performance metric of the edge, wherein the first performance metric and the second performance metric are aggregated; and

smoothing of the performance metrics to account for packet loss and latency associated with the probe packet, wherein the smoothing of the performance metrics comprises estimating the performance metrics of normal traffic as opposed to simulated traffic that includes the probe packet, and wherein the smoothing of the performance metrics comprises grouping at least two of the multiple paths that follow a same path to improve an accuracy of the estimation of the performance metrics,

wherein the endpoint packet loss of the paths is calculated using the following equation: PL=SUM(LP)/SUM(SP) wherein PL represents the packet loss, LP represents the number of packets that were sent to the endpoint and for which no response is received, and SP represents the number of probing packets sent to the endpoint.

2. The method according to claim 1 , wherein the performance metrics comprise at least one of a packet loss or a latency at the two nodes.

3. The method according to claim 1 , wherein the first performance metric and the second performance metric are calculated on a same edge of two different paths or the one path of the multiple discovered paths.

4. The method according to claim 1 , wherein the first performance metrics and the second performance metrics are calculated using a different number of sent probe packets.

5. The method according to claim 1 , wherein the two nodes are consecutive hops on a path.

6. The method according to claim 1 , further comprising:

calculating an edge packet loss based on the packet loss at each of the two nodes.

7. The method according to claim 1 , further comprising:

measuring an estimation error at each of the two nodes; and

using the estimation error to calculate the performance metrics.

8. The method according to claim 1 , wherein a traceroute probe is used to determine the first path and the second path.

9. The method according to claim 1 , wherein the multiple paths are discovered via tracerouting.

10. The method according to claim 1 , wherein the user equipment does not have administrative access to the network.

11. An apparatus, comprising: at least one memory comprising computer program code; at least one processor;

wherein the at least one memory and the computer program code are configured, with the at least one processor, to cause the apparatus at least to:

discover, using a probe packet, multiple paths across a plurality of nodes in a network between the apparatus and an endpoint;

derive, at the apparatus, performance metrics of an edge located between two of the plurality of nodes, wherein the performance metrics are derived based on performance of the two nodes bordering the edge, wherein the two nodes are located within one of the multiple paths;

determine a location of the edge where the performance metrics indicates a problem with the edge between the two nodes;

send an indication to a network provider with the location of the edge, wherein the location is used to signal a problem with one of the two nodes bordering the edge;

calculate the performance metrics based on a first performance metric and a second performance metric of the edge, wherein the first performance metric and the second performance metric are aggregated; and

smooth the performance metrics to account for packet loss and latency associated with a probing packet, wherein the smoothing of the performance metrics comprises estimating, the performance metrics of normal traffic as opposed to simulated traffic that includes the probing packet, and wherein the smoothing of the performance metrics comprises grouping at least two of the multiple paths that follow a same path to improve an accuracy of the estimation of the performance metrics,

wherein the endpoint packet loss of the paths is calculated using the following equation: PL=SUM(LP)/SUM(SP) wherein PL represents the packet loss, LP represents the number of packets that were sent to the endpoint and for which no response is received, and SP represents the number of probing packets sent to the endpoint.

12. The apparatus according to claim 11 , wherein the performance metrics comprise at least one of a packet loss or a latency at the two nodes.

13. The apparatus according to claim 11 , wherein the first performance metric and the second performance metric are calculated on a same edge of two different paths or the one path of the multiple discovered paths.

14. The apparatus according to claim 11 , wherein the first performance metrics and the second performance metrics are calculated using a different number of sent probe packets.

15. The apparatus according to claim 11 , wherein the at least one memory and the computer program code are configured, with the at least one processor, to cause the apparatus at least to:

measure an estimation error at each of the two nodes; and

use the estimation error to calculate the performance metrics.

16. The method according to claim 11 , wherein the user equipment does not have administrative access to the network.

17. A non-transitory computer-readable medium encoding instructions that, when executed in hardware, perform a process, comprising:

discovering, using a probe packet, multiple paths across a plurality of nodes in a network between a user equipment and an endpoint;

deriving, at the user equipment, performance metrics of an edge located between two of the plurality of nodes, wherein the performance metrics are derived based on performance of the two nodes bordering the edge, wherein the two nodes are located within one of the multiple paths;

determining a location of the edge where the performance metrics indicates a problem with the edge;

sending an indication to a network provider with the location of the edge, wherein the location is used to signal a problem with one of the two nodes bordering the edge;

calculating the performance metrics based on a first performance metric and a second performance metric of the edge, wherein the first performance metric and the second performance metric are aggregated; and

smoothing of the performance metrics to account for packet loss and latency associated with a probing packet, wherein the smoothing comprises estimating the performance metrics of normal traffic as opposed to simulated traffic that includes the probing packet, and wherein the smoothing of the performance metrics comprises grouping at least two of the multiple paths that follow a same path to improve an accuracy of the estimation of the performance metrics,

wherein the endpoint packet loss of the paths is calculated using the following equation: PL=SUM(LP)/SUM(SP) wherein PL represents the packet loss, LP represents the number of packets that were sent to the endpoint and for which no response is received, and SP represents the number of probing packets sent to the endpoint.

Assignments (9)

Cited By (1)

US 12,413,471