IP Library Granted Patent US 10,749,711
Granted Patent B2
US 10,749,711 · App. 15/221,608 · Granted Aug 18, 2020

Network-link method useful for a last-mile connectivity in an edge-gateway multipath system

Inventors: Sunil Mukundan (Chennai, IN); Stephen Craig Connors (San Jose, CA); Steven Michael Woo (Los Altos, CA); Ajit Ramachandra Mayya (Saratoga, CA); Thomas Harold Speeter (Gilroy, CA)
Assignee: NICIRA, INC.
H04L12/66H04L1/00H04L1/004H04L1/18H04L43/028H04L45/24H04L47/125H04L47/2408H04L47/41H04L67/02H04L1/1671H04L43/0894
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 10,749,711
App. No.
15/221,608
Granted
Aug 18, 2020
Kind
B2
Abstract

In one exemplary aspect, an edge-gateway multipath method includes the step of providing an edge device in a local network communicatively coupled with a cloud-computing service in a cloud-computing network. A set of wide area network (WAN) links connected to the edge device are automatically detected. The WAN links are automatically measured without the need for an external router. The edge device is communicatively coupled with a central configuration point in the cloud-computing network. The method further includes the step of downloading, from the central configuration point, an enterprise-specific configuration data into the edge device. The enterprise-specific configuration data includes the gateway information. The edge device is communicatively coupled with a gateway in the cloud-computing network. The communicatively coupling of the edge device with the gateway includes a multipath (MP) protocol.

Claims (32)

1. A network-link method useful for establishing connectivity in an edge-gateway multipath, the method comprising:

at an edge device connected to a gateway that is located in a public cloud datacenter:

using, for a network-traffic flow of a computer network, deep-packet inspection to determine an identity of an application type associated with the network-traffic flow;

assessing bandwidth on a set of network links connecting the edge device to the gateway in the public cloud; and

based on the application type identity, intelligently load-balancing traffic on a plurality of the network links by sending successive packets belonging to the network-traffic flow on the plurality of the network links.

2. The network-link method of claim 1 ,

wherein the identity of application type comprises an identity of a specific application associated with the network traffic.

3. The network-link method of claim 2 , wherein a network link comprises a communications channel that connects two or more communicating devices.

4. The network-link methods of claim 3 , wherein the step of intelligently load-balancing the traffic on the plurality of the network links comprises using an application-aware intelligent network link characterization.

5. The network-link method of claim 1 , wherein the application type comprises a real-time application type, a transactional application type, or a bulk-file application type.

6. The network-link method of claim 5 , wherein the network traffic is identified as a bulk file transfer network traffic, and wherein the bulk file transfer network traffic is set as a lowest priority traffic and uses a small portion of a network bandwidth.

7. The network-link method of claim 1 , wherein the application type comprises a social-network website browsing application type, and wherein the network traffic is switched to an internet connection.

8. The method of claim 1 , wherein the plurality of network links comprises a first plurality of links, wherein intelligently load-balancing traffic on the plurality of network links comprises:

identifying a second plurality of active-network links between the edge device and the gateway; and

selecting the first plurality of network links from the second plurality of identified active-network links.

9. The method of claim 8 further comprising:

providing an in-order data delivery with an application persistence by sending data packets belonging to the same traffic flow on the first plurality of active links; and

correcting an error on a lossy network link using an error control mechanism for data transmission selectively based on a current measured condition in the computer network.

10. The network-link method of claim 9 , wherein the error-control mechanism for data transmission comprises an Automatic Repeat-reQuest.

11. The network-link method of claim 9 , wherein the error-control mechanism comprises a forward error correction (FEC).

12. The network-link method of claim 9 , wherein the application type comprises a voice-application type, and wherein the forward-error correction is implemented as the error control mechanism.

13. The method of claim 1 , wherein the plurality of network links includes every link in the set of links.

14. The method of claim 1 , wherein the plurality of network links does not include every link in the set of links.

15. A non-transitory machine readable medium storing a program for establishing connectivity in an edge-gateway multipath, the program for execution by at least one processor of an edge device that connects to a gateway located in a public cloud datacenter, the program comprising sets of instructions for:

using, for a network-traffic flow of a computer network, deep-packet inspection to determine an identity of an application type associated with the network-traffic flow;

assessing bandwidth on a set of network links connecting the edge device to the gateway in the public cloud; and

based on the application type identity, intelligently load-balancing traffic on a plurality of the network links by sending successive packets belonging to the network-flow on the plurality of the network links.

16. The non-transitory machine readable medium of claim 15 , wherein a network link comprises a communications channel that connects two or more communicating devices.

17. The non-transitory machine readable medium of claim 16 , wherein the step of intelligently load-balancing the traffic on the plurality of the network links comprises using an application-aware intelligent network link characterization.

18. The non-transitory machine readable medium of claim 15 , wherein the application type comprises a real-time application type, a transactional application type, or a bulk-file application type.

19. The non-transitory machine readable medium of claim 18 , wherein the network traffic is identified as a bulk file transfer network traffic, and wherein the bulk file transfer network traffic is set as a lowest priority traffic and uses a small portion of a network bandwidth.

20. The non-transitory machine readable medium of claim 15 , wherein the application type comprises a social-network website browsing application type, and wherein the network traffic is switched to an Internet connection.

Assignments (6)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 5, 2025
From: VMWARE, LLC
To: VELOCLOUD NETWORKS, LLC
Reel/Frame 072326/0693 →
MERGER Recorded Jan 27, 2025
From: NICIRA, INC.
To: VMWARE LLC
Reel/Frame 070187/0487 →
CORRECTIVE ASSIGNMENT TO CORRECT THE INADVERTENT DISCREPANCIES IN THE ASSIGNMENT PREVIOUSLY RECORDED ON REEL 044917 FRAME 0102. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT OF ASSIGNOR'S INTEREST. Recorded Jun 18, 2018
From: VELOCLOUD NETWORKS, LLC
To: NICIRA, INC.
Reel/Frame 046129/0882 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 13, 2018
From: VELOCLOUD NETWORKS, LLC
To: NICIRA, INC.
Reel/Frame 044917/0102 →
CHANGE OF NAME Recorded Jan 30, 2018
From: VELOCLOUD NETWORKS, INC.
To: VELOCLOUD NETWORKS, LLC
Reel/Frame 045195/0741 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 26, 2017
From: MAYYA, AJIT RAMACHANDRA; CONNORS, STEPHEN CRAIG; WOO, STEVEN MICHAEL; MUKUNDAN, SUNIL; SPEETER, THOMAS HAROLD
To: VELOCLOUD NETWORKS, INC.
Reel/Frame 043106/0307 →
Continuity (3)
Continuation In Part 14321818 · Jul 2, 2014
Provisional Application 61844822 · Jul 10, 2013
Related Publication 20170134186A1 · May 11, 2017
Cited By (33)
US 12,218,800 US 12,218,845 US 12,237,990 US 12,250,114 US 12,261,777 US 12,267,364 US 12,316,524 US 12,335,131 US 12,355,655 US 12,368,676 US 12,375,403 US 12,401,544 US 12,425,332 US 12,425,335 US 12,425,347 US 12,425,395 US 12,483,968 US 12,489,672 US 12,506,678 US 12,507,120 US 12,507,148 US 12,507,153 US 12,526,183 US 12,549,465 US 12,563,438 US 12,568,039 US 12,587,468 US 12,603,827 US 12,603,848 US 12,632,330 US 12,652,217 US 12,659,719 US 12,719,782