Task offloading and routing in mobile edge cloud networks

Patent metadata

Field	Value
Publication number	US-11503113-B2
Application number	US-202117147972-A
Country	US
Kind code	B2
Filing date	Jan 13, 2021
Priority date	Jul 27, 2018
Publication date	Nov 15, 2022
Grant date	Nov 15, 2022

Abstract

A method implemented by a network element (NE) in a mobile edge cloud (MEC) network, comprising receiving, by the NE, an offloading request message from a client, the offloading request message comprising task-related data describing a task associated with an application executable at the client, determining, by the NE, whether to offload the task to an edge cloud server of a plurality of edge cloud servers distributed within the MEC network based on the task-related data and server data associated with each of the plurality of edge cloud servers, transmitting, by the NE, a response message to the client based on whether the task is offloaded to the edge cloud server.

First claim

What is claimed is:

1. A method implemented by a network element (NE) in a mobile edge cloud (MEC) network, comprising:
receiving an offloading request message from a client, the offloading request message comprising task-related data describing a task associated with an application executable at the client;
determining whether to offload the task to an edge cloud server of a plurality of edge cloud servers distributed within the MEC network based on the task-related data and server data associated with each of the plurality of edge cloud servers by comparing a latency experienced during offloading of the task to the edge cloud server with a latency experienced during local execution of the task at the client, wherein the latency experienced during offloading of the task to the edge cloud server includes a processing delay of one or more messages occurring along a path between the client and the edge cloud server before a result of the task is received by the client, a transmission delay of the one or more messages at one or more NEs on the path, and an execution time that the task remains at the edge cloud server for executing before the result is transmitted by the edge cloud server; and
transmitting a response message to the client based on whether the task is offloaded to the edge cloud server.

2. The method of claim 1 , wherein the offloading request message comprises a block field, wherein the block field is configured to carry metadata associated with the task and instructions to be executed by one or more NEs on a path between the client and the edge cloud server, wherein the metadata comprises at least one of the task-related data, characteristics of the client, or characteristics of the edge cloud server.

3. The method of claim 1 , wherein determining whether to offload the task to the edge cloud server comprises:
comparing an amount of energy consumed by the client during offloading of the task to the edge cloud server with an amount of energy consumed during local execution of the task at the client.

4. The method of claim 3 , wherein comparing the amount of energy consumed during offloading of the task to the edge cloud server with the amount of energy consumed during local execution of the task at the client comprises:
determining an amount of energy consumed by the client during transmission of offloading data, reception of one or more response messages, and waiting for task completion during the offloading of the task; and
determining the amount of energy consumed during local execution of the task at the client.

5. The method of claim 1 , wherein comparing the latency experienced during offloading of the task to the edge cloud server with the latency experienced during local execution of the task at the client comprises:
determining a processing delay of a plurality of messages occurring at a plurality of NEs along a path between the client and the edge cloud server before a result of the task after being executed by the edge cloud server is transmitted to the client;
determining a transmission delay of the plurality of messages forwarded by the plurality of NEs;
determining an execution time for the edge cloud server to perform the task and transmit the result of the task back to the client;
determining the latency experienced during offloading of the task to the edge cloud server based on a sum of the processing delay, transmission delay, and execution time; and
determining the latency experienced during local execution of the task at the client.

6. The method of claim 1 , wherein determining whether to offload the task to the edge cloud server comprises determining that the task should be offloaded to the edge cloud server in response to an amount of energy consumed at the NE during offloading of the task to the edge cloud server being less than an amount of energy consumed during local execution of the task at the client, the latency experienced during offloading of the task to the edge cloud server being less than the latency experienced during local execution of the task at the client, and a response time between the client sending the offloading request message and receiving the response message during offloading of the task to the edge cloud server being less than a deadline response time specified by the client.

7. The method of claim 6 , further comprising determining the edge cloud server from the plurality of edge cloud servers based on an offloading metric associated with each of the plurality of edge cloud servers, wherein the offloading metric is based on a latency and an energy consumption experienced as a result of offloading the task to the edge cloud server.

8. The method of claim 1 , wherein determining whether to offload the task to the edge cloud server comprises determining that the task should be executed locally at the client in response to an amount of energy consumed at the NE during offloading of the task to the edge cloud server being greater than or equal to an amount of energy consumed during local execution of the task at the client.

9. A network element (NE), comprising:
a memory comprising instructions; and
a processor coupled to the memory and configured to execute the instructions, which cause the processor to be configured to:
receive an offloading request message from a client, the offloading request message comprising task-related data describing a task associated with an application executable at the client;
determine whether to offload the task to an edge cloud server of a plurality of edge cloud servers distributed within a network based on the task-related data and server data associated with each of the plurality of edge cloud servers, wherein to determine whether to offload the task to the edge cloud server, the instructions further cause the processor to be configured to compare a latency experienced during offloading of the task to the edge cloud server with a latency experienced during local execution of the task at the client, wherein the latency experienced during offloading of the task to the edge cloud server includes a processing delay of one or more messages occurring along a path between the client and the edge cloud server before a result of the task is received by the client, a transmission delay of the one or more messages at one or more NEs on the path, and an execution time that the task remains at the edge cloud server for executing before the result is transmitted by the edge cloud server; and
transmit a response message to the client based on whether the task is offloaded to the edge cloud server.

10. The NE of claim 9 , wherein the offloading request message comprises a block field, wherein the block field is configured to carry metadata associated with the task and instructions to be executed by one or more NEs on a path between the client and the edge cloud server, wherein the metadata comprises at least one of the task-related data, characteristics of the client, or characteristics of the edge cloud server.

11. The NE of claim 10 , wherein:
to compare an amount of energy consumed during offloading of the task to the edge cloud server with the amount of energy consumed during local execution of the task at the client, the instructions further cause the processor to be configured to:
determine an amount of energy consumed by the client during transmission of offloading data, reception of one or more response messages, and waiting for task completion during the offloading of the task;
determine the amount of energy consumed during local execution of the task at the client; and

to compare the latency experienced during offloading of the task to the edge cloud server with the latency experienced during loca

Assignees

• Huawei Tech Co Ltd

Inventors

• Dong Lijun

Classifications

• H04L67/1008

  based on parameters of servers, e.g. available memory or workload (monitoring of computer activity G06F11/30) · CPC title

• G06F9/5033

  considering data affinity · CPC title

• G06F9/5094

  where the allocation takes into account power or heat criteria (power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206) · CPC title

• H04W4/50

  Service provisioning or reconfiguring · CPC title

• G06F2209/509

  Offload · CPC title