WO2014101534A1

WO2014101534A1 - Profit maximization scheduling method and system for cloud computing

Info

Publication number: WO2014101534A1
Application number: PCT/CN2013/085450
Authority: WO
Inventors: 朱定局
Original assignee: 深圳先进技术研究院
Priority date: 2012-12-25
Filing date: 2013-10-18
Publication date: 2014-07-03
Also published as: CN103002053A; CN103002053B

Abstract

A scheduling method and system for cloud computing. The method comprises: obtaining a plurality of resource data, respectively computing profits of the plurality of resource data according to the plurality of resource data, comparing profits corresponding to the plurality of resource data, so as to obtain a preferential resource datum corresponding to the maximum profit, and performing scheduling on cloud computing resources according to the preferential resource datum. According to the scheduling method and system for cloud computing of the present invention, the profits of different scheduling solutions of the cloud computing resources are computed by subtracting costs from earnings, and a cloud computing resource scheduling solution with the maximum profit is obtained and implemented by comparing the profits.

Description

Cloud computing profit maximization scheduling method and system

[Technical Field]

The present invention relates to cloud computing technologies, and more particularly to a scheduling solution and system for cloud computing technologies.

【Background technique】

In the existing cloud computing technology, when resource scheduling is performed, load balancing of cloud computing resources is mainly considered. However, in the practical application of cloud computing, most users want to obtain the maximum profit through the scheduling of cloud computing resources, especially in the commercial application of cloud computing, the profit of application resources becomes an important reference factor.

Therefore, the existing cloud computing technology aims to pursue load balancing, and such a cloud computing resource scheduling method cannot meet the needs of many commercial application users.

[Summary of the Invention]

Based on this, it is necessary to provide a profit-optimized cloud computing scheduling method and system for the scheduling problem of cloud computing resources. By calculating the profit of various cloud computing scheduling schemes and selecting a profitable scheduling scheme, the cloud computing is implemented. The scheduling of resources to get the most profit.

A scheduling method for cloud computing, comprising the steps of: obtaining a step of obtaining a plurality of resource data; calculating a step of separately calculating profit of the plurality of resource data according to the plurality of resource data; comparing the steps of comparing the profits corresponding to the plurality of resource data Size, obtaining the preferred resource data corresponding to the maximum profit; and scheduling step of scheduling the cloud computing resource according to the preferred resource data.

In one of the embodiments, in the obtaining step, the resource data includes revenue and cost; in the calculating step, the profit is equal to the revenue minus the cost.

In one of the embodiments, the revenue is equal to the amount of tasks completed, and the cost is equal to the number of cloud computing nodes that need to be used.

In one of the embodiments, the revenue is equal to the completed task amount multiplied by the weight k1, and the cost is equal to the number of cloud computing nodes to be used multiplied by the weight K2.

In one embodiment, the weight K1 and the weight K2 are taken from the training database; K1 is the revenue of the unit task, from the task attribute library; K2 is the cost of the unit cloud computing node, from the cloud computing resource attribute library.

A cloud computing scheduling system includes: obtaining a module to obtain a plurality of resource data; a computing module, respectively calculating profit of the plurality of resource data according to the plurality of resource data; comparing the modules, comparing the profit corresponding to the plurality of resource data, The preferred resource data corresponding to the maximum profit is obtained; and the scheduling module performs scheduling of the cloud computing resource according to the preferred resource data.

In one of the embodiments, the resource data obtained by the obtaining module includes revenue and cost; the computing module performs the calculation that the profit is equal to the revenue minus the cost.

The cloud computing scheduling method and system of the present invention calculates the profit of different scheduling schemes of cloud computing resources by reducing the cost, and then obtains and implements the cloud computing resource scheduling scheme with the largest profit by comparing the profit size.

[Description of the Drawings]

1 is a flowchart of a method for scheduling cloud computing according to the present invention;

2 is a schematic block diagram of a scheduling system for cloud computing according to the present invention.

【detailed description】

The invention utilizes the existing cloud computing resources for scheduling, and implements cloud computing with the goal of obtaining the maximum computing profit as much as possible.

As shown in FIG. 1 , the scheduling method of the cloud computing of the present invention includes the following steps:

S1: Obtain a step to obtain multiple resource data. Resource data includes revenue and cost; income and cost have two calculation methods. The first is that the revenue is equal to the completed task quantity, the cost is equal to the number of cloud computing nodes to be used, and the second is that the revenue is equal to the completed task quantity multiplication. With the weight k1, the cost is equal to the number of cloud computing nodes to be used multiplied by the weight K2, and the weight K1 and the weight K2 are taken from the relevant industry training database according to different application industries; K1 is the revenue of the unit task, from the task attribute library; K2 is The cost of a unit cloud computing node comes from the cloud computing resource attribute library.

S2: a calculation step of separately calculating profit of a plurality of resource data according to the plurality of resource data, and the profit is equal to the income minus the cost.

S3: a comparison step of comparing the size of the profit corresponding to the plurality of resource data to obtain the preferred resource data corresponding to the maximum profit.

S4: a scheduling step of scheduling cloud computing resources according to the preferred resource data.

As shown in FIG. 2, the scheduling system of the cloud computing of the present invention includes an obtaining module, a computing module, a comparing module, and a scheduling module that are sequentially connected.

The acquisition module obtains multiple resource data, including revenue and cost. There are two ways to calculate the benefits and costs. The first is that the revenue is equal to the amount of tasks completed, the cost is equal to the number of cloud computing nodes to be used, and the second is that the revenue is equal to the completed task multiplied by the weight k1, and the cost is equal to the need. The number of cloud computing nodes used is multiplied by the weight K2, and the weight K1 and the weight K2 are taken from the relevant industry training database according to different application industries; K1 is the revenue of the unit task, from the task attribute library; K2 is the cost of the unit cloud computing node, From the cloud computing resource property library.

The calculation module separately calculates the profit of the plurality of resource data according to the plurality of resource data.

The comparison module compares the size of the profit corresponding to the plurality of resource data, and obtains the preferred resource data corresponding to the maximum profit.

The scheduling module performs scheduling of the cloud computing resources according to the preferred resource data.

The above-mentioned embodiments are merely illustrative of several embodiments of the present invention, and the description thereof is more specific and detailed, but is not to be construed as limiting the scope of the invention. It should be noted that a number of variations and modifications may be made by those skilled in the art without departing from the spirit and scope of the invention. Therefore, the scope of the invention should be determined by the appended claims.

Claims

A cloud computing scheduling method includes the following steps:

Obtaining steps to obtain multiple resource data;

a calculating step of separately calculating profit of the plurality of resource data according to the plurality of resource data;

a comparison step of comparing the size of the profit corresponding to the plurality of resource data to obtain the preferred resource data corresponding to the maximum profit;

The scheduling step is to perform scheduling of the cloud computing resource according to the preferred resource data.
The method for scheduling cloud computing according to claim 1, wherein in the obtaining step, the resource data includes revenue and cost; and in the calculating step, the profit is equal to the revenue minus the cost.
The cloud computing scheduling method according to claim 2, wherein the revenue is equal to the completed task amount, and the cost is equal to the number of cloud computing nodes to be used.
The cloud computing scheduling method according to claim 2, wherein the revenue is equal to the completed task amount multiplied by a weight k1, the cost being equal to the number of cloud computing nodes to be used multiplied by the weight K2.
The method for scheduling cloud computing according to claim 4, wherein the weight K1 and the weight K2 are taken from a training database; K1 is a revenue of a unit task, from a task attribute library; K2 is a cost of a unit cloud computing node, From the cloud computing resource property library.
A cloud computing scheduling system, comprising:

Obtain a module to obtain multiple resource data;

a calculating module, respectively calculating a profit of the plurality of resource data according to the plurality of resource data;

Comparing a module, comparing the size of the profit corresponding to the plurality of resource data, and obtaining the preferred resource data corresponding to the maximum profit;

The scheduling module performs scheduling of the cloud computing resources according to the preferred resource data.
The scheduling system for cloud computing according to claim 6, wherein the resource data obtained by the obtaining module includes revenue and cost; and the calculating module performs the following calculation: the profit is equal to the revenue minus the cost. .
The cloud computing scheduling system according to claim 7, wherein the revenue is equal to the completed task amount, and the cost is equal to the number of cloud computing nodes to be used.
The cloud computing scheduling system according to claim 7, wherein the revenue is equal to the completed task amount multiplied by a weight k1, the cost being equal to the number of cloud computing nodes to be used multiplied by the weight K2.
The scheduling system for cloud computing according to claim 9, wherein the weight K1 and the weight K2 are taken from a training database; K1 is a revenue of a unit task, from a task attribute library; K2 is a cost of a unit cloud computing node, From the cloud computing resource property library.