IP Library Granted Patent US 12,470,472
Granted Patent B2
US 12,470,472 · App. 18/244,597 · Granted Nov 11, 2025

Optimized agent deployment for datacenter monitoring

Inventors: Raghu Rajendra Arur (New Delhi, IN); Indermeet Singh Gandhi (San Jose, CA)
Assignee: Cisco Technology, Inc.
H04L43/50H04L43/065H04L43/0852
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,470,472
App. No.
18/244,597
Granted
Nov 11, 2025
Kind
B2
Abstract

In one implementation, a device determines a plurality of routes from one or more sources in a localized network domain to external addresses via a plurality of border leaf nodes of the localized network domain. The device selects, based on a specific performance test from a particular source of the one or more sources to a particular endpoint with a corresponding external address, a particular border leaf node of the plurality of border leaf nodes that is along a utilized route from the particular source to the particular endpoint based on the corresponding external address and the plurality of routes. The device causes the specific performance test to be executed from a particular testing agent installed on the particular border leaf node to the particular endpoint.

Claims (38)

1 . A method, comprising:

determining, by a device, a plurality of routes from one or more sources in a localized network domain to external addresses via a plurality of border leaf nodes of the localized network domain, wherein each of the plurality of border leaf nodes connects the localized network domain to an external network;

selecting, by the device and based on a specific performance test from a particular source of the one or more sources to a particular endpoint with a corresponding external address, a particular border leaf node of the plurality of border leaf nodes that is along a utilized route from the particular source to the particular endpoint based on the corresponding external address and the plurality of routes;

causing, by the device, the specific performance test to be executed from a particular testing agent installed on the particular border leaf node to the particular endpoint;

stitching, by the device and into end-to-end path metrics, performance metrics of the localized network domain from the particular source to the particular border leaf node to results of the specific performance test from the particular testing agent installed on the particular border leaf node to the particular endpoint; and

correlating, by the device, the end-to-end path metrics to the particular source as a result of the specific performance test from the particular source to the particular endpoint.

2 . The method as in claim 1 , further comprising:

receiving a request to initiate the specific performance test from a network performance testing agent controller, wherein causing comprises informing the network performance testing agent controller of the particular border leaf node that is along the utilized route from the particular source to the particular endpoint.

3 . The method as in claim 1 , wherein the specific performance test is configured to deem the performance metrics of the localized network domain negligible.

4 . The method as in claim 1 , wherein the performance metrics of the localized network domain comprise one or more of latency, buffer drops, or jitter.

5 . The method as in claim 1 , wherein the performance metrics of the localized network domain are based on network monitoring unrelated to the particular testing agent.

6 . The method as in claim 1 , wherein the localized network domain comprises a datacenter.

7 . The method as in claim 1 , wherein the one or more sources comprise one or more of servers or virtual machines.

8 . The method as in claim 1 , wherein determining the plurality of routes is based on virtual routing and forwarding tables within the localized network domain and border gateway protocol sessions of the plurality of border leaf nodes.

9 . The method as in claim 1 , determining the plurality of routes is based on one or more access control lists.

10 . The method as in claim 1 , wherein the device comprises a traffic performance aggregator.

11 . The method as in claim 1 , wherein the particular endpoint comprises a service with one or more corresponding external addresses.

12 . The method as in claim 11 , wherein the service is associated with a plurality of corresponding external addresses, the method further comprising:

selecting a plurality of particular border leaf nodes that are along a plurality of utilized routes from the particular source to the particular endpoint based on the plurality of corresponding external addresses and the plurality of routes; and

causing the specific performance test to be executed from a plurality of particular testing agents installed respectively on the plurality of particular border leaf nodes.

13 . A tangible, non-transitory, computer-readable medium having computer-executable instructions stored thereon that, when executed by a processor on a computer, cause the computer to perform a method comprising:

determining a plurality of routes from one or more sources in a localized network domain to external addresses via a plurality of border leaf nodes of the localized network domain, wherein each of the plurality of border leaf nodes connect the localized network domain to an external network;

selecting, based on a specific performance test from a particular source of the one or more sources to a particular endpoint with a corresponding external address, a particular border leaf node of the plurality of border leaf nodes that is along a utilized route from the particular source to the particular endpoint based on the corresponding external address and the plurality of routes;

causing the specific performance test to be executed from a particular testing agent installed on the particular border leaf node to the particular endpoint;

stitching, into end-to-end path metrics, performance metrics of the localized network domain from the particular source to the particular border leaf node to results of the specific performance test from the particular testing agent installed on the particular border leaf node to the particular endpoint; and

correlating the end-to-end path metrics to the particular source as a result of the specific performance test from the particular source to the particular endpoint.

14 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein the method further comprises:

receiving a request to initiate the specific performance test from a network performance testing agent controller, wherein causing comprises informing the network performance testing agent controller of the particular border leaf node that is along the utilized route from the particular source to the particular endpoint.

15 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein the specific performance test is configured to deem the performance metrics of the localized network domain negligible.

16 . An apparatus, comprising:

one or more network interfaces to communicate with a network;

a processor coupled to the one or more network interfaces and configured to execute one or more processes; and

a memory configured to store a process that is executable by the processor, the process, when executed, configured to:

determine a plurality of routes from one or more sources in a localized network domain to external addresses via a plurality of border leaf nodes of the localized network domain, wherein each of the plurality of border leaf nodes connects the localized network domain to an external network;

select, based on a specific performance test from a particular source of the one or more sources to a particular endpoint with a corresponding external address, a particular border leaf node of the plurality of border leaf nodes that is along a utilized route from the particular source to the particular endpoint based on the corresponding external address and the plurality of routes;

cause the specific performance test to be executed from a particular testing agent installed on the particular border leaf node to the particular endpoint;

stitch, into end-to-end path metrics, performance metrics of the localized network domain from the particular source to the particular border leaf node to results of the specific performance test from the particular testing agent installed on the particular border leaf node to the particular endpoint; and

correlate the end-to-end path metrics to the particular source as a result of the specific performance test from the particular source to the particular endpoint.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 11, 2023
From: ARUR, RAGHU RAJENDRA; GANDHI, INDERMEET SINGH
To: CISCO TECHNOLOGY, INC.
Reel/Frame 064861/0853 →
Continuity (1)
Related Publication 20250088448A1 · Mar 13, 2025
References Cited (24)
US 6408335B1 · Schwaller · 2002 [cited by examiner]
US 8995301B1 · Miller · 2015 [cited by examiner]
US 10819556B1 · Rangasamy et al. · 2020 [cited by applicant]
US 11271775B2 · Skidmore et al. · 2022 [cited by applicant]
US 11444871B1 · Nainar · 2022 [cited by examiner]
US 11539735B2 · Rao et al. · 2022 [cited by applicant]
US 20070165532A1 · Retana · 2007 [cited by examiner]
US 20090268713A1 · Ottur · 2009 [cited by examiner]
US 20120027016A1 · Filsfils · 2012 [cited by examiner]
US 20130019005A1 · Hui · 2013 [cited by examiner]
US 20140222748A1 · Mermoud · 2014 [cited by examiner]
US 20160359695A1 · Yadav et al. · 2016 [cited by applicant]
US 20170078176A1 · Lakshmikantha · 2017 [cited by examiner]
US 20170126647A1 · Zhang · 2017 [cited by examiner]
US 20170329696A1 · Bitoun · 2017 [cited by examiner]
US 20180152861A1 · Rodriguez et al. · 2018 [cited by applicant]
US 20180219899A1 · Joy · 2018 [cited by examiner]
US 20200106631A1 · Bosch · 2020 [cited by examiner]
US 20200136973A1 · Rahman · 2020 [cited by examiner]
US 20200379839A1 · Savalle · 2020 [cited by examiner]
US 20210288898A1 · Shen · 2021 [cited by examiner]
US 20220046045A1 · Rao · 2022 [cited by examiner]
US 20220321457A1 · Hari · 2022 [cited by examiner]
“NVIDIA Cumulus Linux—Network Reference Design Guide”, 2022, 43 pages, NVIDIA Corporation. [cited by applicant]