Traffic optimization for multi-node applications

Patent metadata

Field	Value
Publication number	US-2018295064-A1
Application number	US-201715484049-A
Country	US
Kind code	A1
Filing date	Apr 10, 2017
Priority date	Apr 10, 2017
Publication date	Oct 11, 2018
Grant date	—

Abstract

Example methods are provided for a host to perform traffic optimization for a multi-node application in a virtualized computing environment. The method may comprise: detecting, from the first virtualized computing instance supported by the host, a packet that includes header information identifying the second virtualized computing instance as a destination; and identifying that the first virtualized computing instance and second virtualized computing instance are configured to perform respective first role and second role to implement the multi-node application. The method may also comprise: based on one or more optimization factors, selecting the third virtualized computing instance that is also configured to perform the second role to replace the second virtualized computing instance; modifying the header information of the packet to identify the third virtualized computing instance as the destination; and sending the packet with modified header information to the third virtualized computing instance.

First claim

I claim:

1 . A method for a host to perform traffic optimization for a multi-node application implemented by at least a first virtualized computing instance, a second virtualized computing instance and a third virtualized computing instance in a virtualized computing environment, wherein the method comprises:
detecting, from the first virtualized computing instance supported by the host, a packet that includes header information identifying the second virtualized computing instance as a destination; identifying that the first virtualized computing instance and second virtualized computing instance are configured to perform respective first role and second role to implement the multi-node application; based on one or more optimization factors, selecting the third virtualized computing instance that is also configured to perform the second role to replace the second virtualized computing instance; modifying the header information of the packet to identify the third virtualized computing instance as the destination; and sending the packet with modified header information to the third virtualized computing instance.

2 . The method of claim 1 , wherein selecting the third virtualized computing instance comprises:
based on a location-based optimization factor, determining that the first virtualized computing instance and third virtualized computing instance are both associated with a first location and the second virtualized computing instance is associated with a second location, wherein the first location and second location represent different hosts, racks, pods, chassis, data centers, or clouds.

3 . The method of claim 1 , wherein selecting the third virtualized computing instance comprises:
selecting the third virtualized computing instance based on performance-based optimization factors associated with one or more of the following: central processing unit (CPU) resource utilization, memory resource utilization, storage resource utilization, network resource utilization, and error performance.

4 . The method of claim 1 , wherein the method further comprises:
prior to selecting the third virtualized computing instance, determining that the first virtualized computing instance is not exclusively associated with the second virtualized computing instance.

5 . The method of claim 1 , wherein the method further comprises:
prior to detecting the packet, obtaining role mapping data that associates the first virtualized computing instance with the first role, and the second virtualized computing instance and third virtualized computing instance with the second role.

6 . The method of claim 1 , wherein modifying the header information comprises:
generating a network address translation (DNAT) rule that associates a second network address of the second virtualized computing instance with a third network address of the third virtualized computing instance; and based on the DNAT rule, replacing the second network address in the header information with the third network address such that the packet is addressed from a first network address of the first virtualized computing instance to the third network address.

7 . The method of claim 6 , wherein the method further comprises:
detecting a response packet addressed from the third network address to the first network address; based on the DNAT rule, modifying header information of the response packet by replacing the third network address with the second network address; and sending the response packet with the modified header information to the first virtualized computing instance.

8 . A non-transitory computer-readable storage medium that includes a set of instructions which, in response to execution by a processor of a host, cause the processor to perform a method of traffic optimization for a multi-node application implemented by at least a first virtualized computing instance, a second virtualized computing instance and a third virtualized computing instance in a virtualized computing environment, wherein the method comprises:
detecting, from the first virtualized computing instance supported by the host, a packet that includes header information identifying the second virtualized computing instance as a destination; identifying that the first virtualized computing instance and second virtualized computing instance are configured to perform respective first role and second role to implement the multi-node application; based on one or more optimization factors, selecting the third virtualized computing instance that is also configured to perform the second role to replace the second virtualized computing instance; modifying the header information of the packet to identify the third virtualized computing instance as the destination; and sending the packet with modified header information to the third virtualized computing instance.

9 . The non-transitory computer-readable storage medium of claim 8 , wherein selecting the third virtualized computing instance comprises:
based on a location-based optimization factor, determining that the first virtualized computing instance and third virtualized computing instance are both associated with a first location and the second virtualized computing instance is associated with a second location, wherein the first location and second location represent different hosts, racks, pods, chassis, data centers, or clouds.

10 . The non-transitory computer-readable storage medium of claim 8 , wherein selecting the third virtualized computing instance comprises:
selecting the third virtualized computing instance based on performance-based optimization factors associated with one or more of the following: central processing unit (CPU) resource utilization, memory resource utilization, storage resource utilization, network resource utilization, and error performance.

11 . The non-transitory computer-readable storage medium of claim 8 , wherein the method further comprises:
prior to selecting the third virtualized computing instance, determining that the first virtualized computing instance is not exclusively associated with the second virtualized computing instance.

12 . The non-transitory computer-readable storage medium of claim 8 , wherein the method further comprises:
prior to detecting the packet, obtaining role mapping data that associates the first virtualized computing instance with the first role, and the second virtualized computing instance and third virtualized computing instance with the second role.

13 . The non-transitory computer-readable storage medium of claim 8 , wherein modifying the header information comprises:
generating a network address translation (DNAT) rule that associates a second network address of the second virtualized computing instance with a third network address of the third virtualized computing instance; and based on the DNAT rule, replacing the second network address in the header information with the third network address such that the packet is addressed from a first network address of the first virtualized computing instance to the third network address.

14 . The non-transitory computer-readable storage medium of claim 13 , wherein the method further comprises:
detecting a response packet addressed from the third network address to the first network address; based on the DNAT rule, modifying header information of the response packet by replacing the third network address with the second network address; and sending the response packet with the modified header information to the first virtualized computing instance.

15 . A host configured to traff

Assignees

• Vmware Inc

Inventors

• Movsisyan Vardan

Classifications

• H04L45/70

  Routing based on monitoring results · CPC title

• H04L61/2521

  Translation architectures other than single NAT servers · CPC title

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title

• H04L69/22 Primary

  Parsing or analysis of headers · CPC title

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title