Service mesh offload to network devices

Patent metadata

Field	Value
Publication number	US-12323482-B2
Application number	US-202117238960-A
Country	US
Kind code	B2
Filing date	Apr 23, 2021
Priority date	Apr 23, 2021
Publication date	Jun 3, 2025
Grant date	Jun 3, 2025

Abstract

Examples described herein relate to a switch comprising a programmable data plane pipeline, wherein the programmable data plane pipeline is configured to provide microservice-to-microservice communications within a service mesh. In some examples, to provide microservice-to-microservice communications within a service mesh, the programmable data plane pipeline is to perform a forwarding operation for a communication from a first microservice to a second microservice. In some examples, to perform a forwarding operation for a communication from a first microservice to a second microservice, the programmable data plane pipeline is to utilize a reliable transport protocol.

First claim

The invention claimed is:

1. A system comprising:
switch circuitry comprising programmable data plane pipeline circuitry, wherein the programmable data plane pipeline circuitry is configured to provide microservice-to-microservice communications within a service mesh;
wherein:
the programmable data plane pipeline circuitry comprises match-action circuitry;
the match-action circuitry comprises match table data and an action engine;
the action engine is to compare header field data of packet data received by the switch circuitry with the match table data to determine one or more actions to be applied to the packet data by the action engine; and
the one or more actions are to implement, at least in part, microservice selection, load balancing, and telemetry processing to be comprised, at least in part, in providing the microservice-to-microservice communications.

2. The system of claim 1 , wherein to provide microservice-to-microservice communications within a service mesh, the programmable data plane pipeline circuitry is to perform a forwarding operation for a communication from a first microservice to a second microservice.

3. The system of claim 2 , wherein to perform a forwarding operation for a communication from a first microservice to a second microservice, the programmable data plane pipeline circuitry is to utilize a reliable transport protocol.

4. The system of claim 1 , wherein the programmable data plane pipeline circuitry is to select a processor, virtual machine, or container to execute an instance of a target microservice based on hardware resource availability.

5. The system of claim 1 , wherein the programmable data plane pipeline circuitry is configured to program at least one programmable data plane pipeline in at least one network device associated with a target microservice within the service mesh, wherein to program at least one programmable data plane pipeline in at least one network device comprises program one or more of: a reliable transport protocol, encryption and decryption scheme, and/or telemetry reporting.

6. The system of claim 1 , wherein the programmable data plane pipeline circuitry is configured to provide microservice-to-microservice communications within a service mesh by communication with a service mesh controller and wherein the communication with a service mesh controller comprises communication between an instance of Istio with an Envoy application program interface (API).

7. A method comprising:
in programmable packet processing pipeline circuitry of at least one of network devices associated with a service mesh of microservices, performing selection of microservices for execution and providing communication within the service mesh of microservices;
wherein:
the programmable packet processing pipeline circuitry comprises match-action circuitry;
the match-action circuitry comprises match table data and an action engine;
the action engine is to compare header field data of received packet data with the match table data to determine one or more actions to be applied to the received packet data by the action engine; and
the one or more actions are to implement, at least in part, (1) the selection of the microservices, and (2) load balancing and telemetry processing to be comprised, at least in part, in the providing of the communication.

8. The method of claim 7 , wherein the network devices comprise a switch that includes the programmable packet processing pipeline circuitry and wherein the programmable packet processing pipeline circuitry of the switch performs microservice load balancing and traffic management.

9. The method of claim 7 , wherein the network devices comprise a network interface controller that includes a programmable packet processing pipeline and wherein the programmable packet processing pipeline of the network interface controller communicates messages between microservices using a reliable transport protocol.

10. The method of claim 7 , wherein the network devices comprise a network interface controller that includes a programmable packet processing pipeline, wherein the programmable packet processing pipeline of the network interface controller is associated with a gateway to a service mesh, and wherein the packet processing pipeline of the network interface controller associated with a gateway performs one service-to-one port binding, implementation of a reliable transport protocol, packet encryption or decryption, and telemetry reporting.

11. The method of claim 7 , comprising:
configuring the programmable packet processing pipeline circuitry by communication with a service mesh controller, wherein the communication with a service mesh controller comprises communication between an instance of Istio with an Envoy application program interface (API).

12. The method of claim 7 , comprising:
receiving a request to perform a service at a gateway among the network devices associated with a service mesh of microservices, wherein the gateway performs protocol translation and provides the request to a switch among the network devices associated with a service mesh of microservices.

13. The method of claim 12 , comprising:
at the switch:
selecting an instance of a microservice to execute;
sending a request to execute the selected microservice; and
receiving a response to the request to execute the selected microservice, wherein the response comprises data to forward or a request to perform a second microservice.

14. The method of claim 7 , wherein the microservices of the mesh of microservices execute on a server, a network device, and/or an infrastructure processing unit (IPU).

15. A non-transitory computer-readable medium storing instructions, that if executed by one or more processors, result in the one or more processors performing operations comprising:
configure a data plane of a switch to:
select a microservice to execute in response to a service request received at a service mesh; and
configure at least one network device associated with the service mesh to apply a reliable transport protocol to communicate with the switch, wherein the configure at least one network device associated with the service mesh to apply a reliable transport protocol to communicate with the switch comprises configure at least one data plane of the at least one network device to apply a reliable transport protocol to communicate with the switch;

wherein:
the data plane comprises programmable data plane pipeline circuitry that comprises match-action circuitry;
the match-action circuitry comprises match table data and an action engine;
the action engine is to compare header field data of packet data received by the switch with the match table data to determine one or more actions to be applied to the packet data by the action engine; and
the one or more actions are to implement, at least in part, microservice selection, load balancing, and telemetry processing to be comprised, at least in part, in providing microservice-to-microservice communications.

16. The computer-readable medium of claim 15 , wherein the switch is to perform load balancing of microservice execution and traffic management within the service mesh.

17. The computer-readable medium of claim 15 , wherein the at least one network device associated with the service mesh comprises a network interface controller and/or an infrastructure processing unit (IPU).

18. The computer-readable medium of claim 15 , wherein the reliable transport protocol to communicate with the switch is based on Address Family of eXpress Data Path (AF_

Assignees

• Intel Corp

Inventors

• He Shaopeng
• Liang Cunming
• Kang Haitao
• Ni Hongjun
• Yu Jiang
• Yang Ziye
• Jain Anjali Singhai
• Daly Daniel
• Li Yadong
• Yu Ping
• Cui Bo
• Wu Jingjing
• Ma Liang
• Liu Changpeng

Classifications

• H04L67/51

  Discovery or management thereof, e.g. service location protocol [SLP] or web services · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L67/63

  Routing a service request depending on the request content or context · CPC title

• H04L67/1001

  for accessing one among a plurality of replicated servers · CPC title

• H04L67/10 Primary

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title