Method and server for dynamic work transfer

Patent metadata

Field	Value
Publication number	US-10764359-B2
Application number	US-201815924297-A
Country	US
Kind code	B2
Filing date	Mar 19, 2018
Priority date	Jan 4, 2018
Publication date	Sep 1, 2020
Grant date	Sep 1, 2020

Abstract

A method and a server for dynamic work transfer are provided. The method includes following steps: regularly collecting and recording network resources of multiple nodes including a cloud node and multiple edge nodes in a network; receiving a request of a first job at a first time point, calculating a cost for configuring the first job to each node according to the network resource of each node at the first time point, and configuring the first job to a first target node; receiving a request of a second job at a second time point, calculating a cost for configuring the second job to each node according to the network resource of each node at the first time point, and determining a second target node suitable for configuring the second job and whether to transfer the first job; and accordingly configuring the second job and maintaining or transferring the first job.

First claim

What is claimed is:

1. A method for dynamic work transfer, adapted for a cloud node under a cloud and edge computing framework, the method comprising:
regularly collecting and recording network resources of a plurality of nodes in a network, wherein the nodes comprise a cloud node and a plurality of edge nodes;
in response to receiving a request of a first job at a first time point, calculating costs for configuring the first job to the respective nodes based on the network resources of the respective nodes at the first time point to configure the first job to a first target node of the nodes;
in response to receiving a request of a second job at a second time point, calculating costs for configuring the second job to the respective nodes based on the network resources of the respective nodes at the first time point, and determining a second target node suitable for configuring the second job and whether to transfer the first job, wherein the second time point is after the first time point; and
configuring the second job and maintaining or transferring the first job based on a determination result.

2. The method for dynamic work transfer as claimed in claim 1 , wherein determining whether to transfer the first job comprises:
determining whether the second target node suitable for configuring the second job is the same as the first target node;
configuring the second job to the second target node without transferring the first job if the second target node suitable for configuring the second job is not the same as the first target node;
determining whether a cost for configuring the second job to the first target node is greater than a cost for configuring the first job to the first target node at the first time point if the second target node suitable for configuring the second job is the same as the first target node;
re-calculating costs for configuring the second job to the respective nodes based on the network resources of the respective nodes at the second time point to configure the second job to a third target node without transferring the first job if the cost for configuring the second job to the first target node is greater than the cost for configuring the first job to the first target node at the first time point; and
transferring the first job to a fourth target node as an adjacent node of the first target node or not transferring the first job based on a cost for configuring the first job to the adjacent node at the first time point and a cost for maintaining the first job at the first target node at the second time point if the cost for configuring the second job to the first target node is not greater than the cost for configuring the first job to the first target node at the first time point.

3. The method for dynamic work transfer as claimed in claim 1 , wherein the network resources comprise respective transmission latencies with respect to other nodes obtained by each of the edge nodes through regularly exchanging messages with the other nodes.

4. The method for dynamic work transfer as claimed in claim 1 , wherein calculating the costs for configuring the first job to the respective nodes based on the network resources of the respective nodes at the first time point comprises:
calculating a transmission cost, a calculation cost, and a storage cost required for configuring the first job to the respective node based on the network resource of each of the nodes at the first time point, and calculating a weighted sum of the transmission cost, the calculation cost, and the storage cost as the cost for configuring the first job to the node.

5. The method for dynamic work transfer as claimed in claim 4 , wherein calculating the transmission cost, the calculation cost, and the storage cost required for configuring the first job to the respective node based on the network resource of each of the nodes at the first time point comprises:
normalizing the network resource of each of the nodes to convert the network resource into the transmission cost, the calculation cost, and the storage cost.

6. The method for dynamic work transfer as claimed in claim 1 , wherein the first job is a job n th previous to the second job, wherein n is a positive integer.

7. The method for dynamic work transfer as claimed in claim 1 , wherein configuring the second job and maintaining or transferring the first job based on the determination result comprises:
only transferring a job configured to a node keeping k hops or fewer from a transmitting end of the second job, wherein k is a positive integer.

8. A server, adapted as a cloud node under a cloud and edge computing framework, the server comprising:
a communication apparatus, connected to a network and communicating with a plurality of edge nodes in the network;
a storage apparatus; and
a processor, coupled to the communication apparatus and the storage apparatus and executing a program recorded in the storage apparatus to:
regularly collect network resources of a plurality of nodes in the network through the communication apparatus and record the network resources in the storage apparatus, wherein the nodes comprise the cloud node and the edge nodes;
in response to receiving a request of a first job through the communication apparatus at a first time point, calculate costs for configuring the first job to the respective nodes based on the network resources of the respective nodes at the first time point to configure the first job to a first target node of the nodes;
in response to receiving a request of a second job at a second time point, calculate costs for configuring the second job to the respective nodes based on the network resources of the respective nodes at the first time point, and determine a second target node suitable for configuring the second job and whether to transfer the first job, wherein the second time point is after the first time point; and
configure the second job and maintain or transfer the first job based on a determination result.

9. The server as claimed in claim 8 , wherein the processor comprises:
determining whether the second target node suitable for configuring the second job is the same as the first target node;
configuring the second job to the second target node without transferring the first job by the processor if the second target node suitable for configuring the second job is not the same as the first target node;
determining whether a cost for configuring the second job to the first target node is greater than a cost for configuring the first job to the first target node at the first time point by the processor if the second target node suitable for configuring the second job is the same as the first target node;
re-calculating costs for configuring the second job to the respective nodes based on the network resources of the respective nodes at the second time point to configure the second job to a third target node without transferring the first job by the processor if the cost for configuring the second job to the first target node is greater than the cost for configuring the first job to the first target node at the first time point; and
transferring the first job to a fourth target node as an adjacent node of the first target node or not transferring the first job by the processor based on a cost for configuring the first job to the adjacent node at the first time point and a cost for maintaining the first job at the first target node at the second time point if the cost for configuring the second job to the first target node is not greater than the cost for configuring the first job to the first target node at the first time point.

10. The server as claimed in claim 8 , wherein the network resources compri

Assignees

• Ind Tech Res Inst

Inventors

• Chen Lyu-Han
• Wang Ching-Yao

Classifications

• 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

• H04L67/563

  Data redirection of data network streams · CPC title

• H04L67/63

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

• H04L43/10

  Active monitoring, e.g. heartbeat, ping or trace-route · CPC title