# US 2017/0286147 A1 · System and method for load estimation of virtual machines in a cloud environment and serving node

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2017/0286147 A1
- Document ID: US20170286147A1
- Status: Published application
- Kind: A1
- Application number: 15/512,191
- Filed: 2015-09-02
- Published: 2017-10-05
- Earliest claimed: 2014-09-22
- Assignees: China Unionpay Co., Ltd.
- Applicants: China Unionpay Co., Ltd.
- Inventors: Hongfeng CHAI; Chengrong WU; Jinzhi HUA; Zhenfang WANG; Yang YANG; Jun ZHANG
- Web page: https://patentgrep.com/patent/US20170286147A1

## CPC

- G06F 9/45558: {Hypervisor-specific management and integration aspects}
- G06F 9/5077: {Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455)}
- G06F 11/3006: {where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50)}
- G06F 11/3433: {for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029)}
- G06F 9/46: Multiprogramming arrangements
- G06F 9/5072: {Grid computing}
- G06F 9/5083: {Techniques for rebalancing the load in a distributed system}
- G06F 11/301: {where the computing system is a virtual computing platform, e.g. logically partitioned systems (virtual machines G06F9/45533; logical partitioning of resources G06F9/5077)}
- G06F 2009/4557: {Distribution of virtual machine instances; Migration and load balancing}
- G06F 2201/815: Virtual
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

The application discloses a load evaluation system for virtual machines in a cloud environment, comprising: a monitored data collecting module which is configured to collect monitored data from one or more virtual machines in a cloud data center; a monitored data processing module which is coupled with the monitored data collecting module and is configured to: perform aggregated grouping on the monitored data by using a first KMeans algorithm for each of the one or more virtual machines, each group having a center point coordinate, and calculate a performance characteristic of each virtual machine of the one or more virtual machines according to the grouping and the center point coordinates of individual groups, i.e., a first characteristic value; perform aggregated grouping on the one or more virtual machines using a second KMeans algorithm based on the first characteristic value of each virtual machine, thus determining the characteristic type and performance load value of each virtual machine of the one or more virtual machines. The application also discloses a load evaluation method for virtual machines in a cloud environment as well as a service node.

## Claims

### Claim 1

A load evaluation system for virtual machines in a cloud environment, comprising:
a monitored data collecting module which is configured to collect monitored data from one or more virtual machines in a cloud data center; and
a monitored data processing module which is coupled with the monitored data collecting module and is configured to:
perform aggregated grouping on the monitored data by using a first KMeans algorithm for each of the one or more virtual machines, each group having a center point coordinate, and calculate a performance characteristic of each virtual machine of the one or more virtual machines according to the grouping and the center point coordinates of individual groups, i.e., a first characteristic value;
perform aggregated grouping on the one or more virtual machines using a second KMeans algorithm based on the first characteristic value of each virtual machine, thus determining the characteristic type and performance load value of each virtual machine of the one or more virtual machines,
wherein a type parameter K in the first KMeans algorithm is larger than a type parameter in the second KMeans algorithm, an initial center of the first KMeans algorithm is selected so that a distance that is as far as possible is kept between initial types, and an initial center of the second KMeans algorithm is selected to be a subset of the initial center of the first KMeans algorithm.

### Claim 2 (depends on 1)

The load evaluation system for virtual machines according to claim 1, wherein the monitored data is associated with a central processing unit, a memory, hard disk input or output, and network input or output of one or more virtual machines.

### Claim 3 (depends on 1)

The load evaluation system for virtual machines according to claim 1, wherein the characteristic type of the virtual machine comprises CPU-intensive type, Network-intensive type and Memory-intensive type.

### Claim 4 (depends on 1)

The load evaluation system for virtual machines according to claim 1, wherein the type parameter K in the first KMeans algorithm is equal to 4, and the type parameter K in the second KMeans algorithm is equal to 3.

### Claim 5 (depends on 4)

The load evaluation system for virtual machines according to claim 4, wherein the initial center of the first KMeans algorithm is selected to be P1 (100,0,0), P2 (0,100,0), P3 (0,0, 100) and P4 (0,0,0).

### Claim 6 (depends on 4)

The load evaluation system for virtual machines according to claim 4, wherein the proportions of samples in each group is A%, B%, C% and D%, and the center coordinates of each group are (X1, Y1, Z1), (X2, Y2, Z2), (X3, Y3, Z3) and (X4, Y4, Z4) respectively, then the first characteristic value is expressed by the following equation:
λ
i
=

(
A


%
,
B


%
,
C


%
,
D


%
)
*
[
X


1
Y


1
Z


1
X


2
Y


2
Z


2
X


3
Y


3
Z


3
X


4
Y


4
Z


4
]

=

(
A


%
*
X


1
+

B


%
*
X


2
+
C


%
*
X


3
+
D


%
*
X


4
,

A


%
*
Y


1
+
B


%
*
Y


2
+
C


%
*
Y


3
+
D


%
*
Y


4
,

A


%
*
Z


1
+
B


%
*
Z


2
+
C


%
*
Z


3
+
D


%
*
Z


4
)

### Claim 7 (depends on 1)

The load evaluation system for virtual machines according to claim 1, wherein the monitored data processing module is further configured to performing redistribution on the virtual machines according to the performance load values in the one or more virtual machines, thereby realizing an effective utilization of resources.

### Claim 8

A load evaluation method for virtual machines in a cloud environment, comprising:
performing aggregated grouping on monitored data obtained from one or more virtual machines in a could data center by using a first KMeans algorithm for each of the one or more virtual machines, each group having a center point coordinate, and calculating a performance characteristic of each virtual machine of the one or more virtual machines according to the grouping and the center point coordinates of individual groups, i.e., a first characteristic value;
performing aggregated grouping on the one or more virtual machines using a second KMeans algorithm based on the first characteristic value of each virtual machine, thus determining the characteristic type and performance load value of each virtual machine of the one or more virtual machines,
wherein a type parameter K in the first KMeans algorithm is larger than a type parameter in the second KMeans algorithm, an initial center of the first KMeans algorithm is selected so that a distance that is as far as possible is kept between initial types, and an initial center of the second KMeans algorithm is selected to be a subset of the initial center of the first KMeans algorithm.

### Claim 9 (depends on 8)

The method according to claim 8, wherein the monitored data is associated with a central processing unit, a memory, hard disk input or output, and network input or output of one or more virtual machines.

### Claim 10 (depends on 8)

The method according to claim 8, wherein the characteristic type of the virtual machine comprises CPU-intensive type, Network-intensive type and Memory-intensive type.

### Claim 11 (depends on 8)

The method according to claim 8, wherein the type parameter K in the first KMeans algorithm is equal to 4, and the type parameter K in the second KMeans algorithm is equal to 3.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the initial center of the first KMeans algorithm is selected to be P1 (100,0,0), P2 (0,100,0), P3 (0,0, 100) and P4 (0,0,0).

### Claim 13 (depends on 11)

The method according to claim 11, wherein the proportions of samples in each group is A%, B%, C% and D%, and the center coordinates of each group are (X1, Y1, Z1), (X2, Y2, Z2), (X3, Y3, Z3) and (X4, Y4, Z4) respectively, then the first characteristic value is expressed by the following equation:
λ
i
=

(
A


%
,
B


%
,
C


%
,
D


%
)
*
[
X


1
Y


1
Z


1
X


2
Y


2
Z


2
X


3
Y


3
Z


3
X


4
Y


4
Z


4
]

=

(
A


%
*
X


1
+

B


%
*
X


2
+
C


%
*
X


3
+
D


%
*
X


4
,

A


%
*
Y


1
+
B


%
*
Y


2
+
C


%
*
Y


3
+
D


%
*
Y


4
,

A


%
*
Z


1
+
B


%
*
Z


2
+
C


%
*
Z


3
+
D


%
*
Z


4
)

### Claim 14 (depends on 8)

The method according to claim 8, further comprising performing redistribution on the virtual machines according to the performance load values in the one or more virtual machines, thereby realizing an effective utilization of resources.

### Claim 15

(canceled)