Method for solving the decomposition-coordination calculation based on block bordered diagonal form (BBDF) model using data center

Patent metadata

Field	Value
Publication number	US-10235341-B2
Application number	US-201414901726-A
Country	US
Kind code	B2
Filing date	Sep 22, 2014
Priority date	Jun 12, 2014
Publication date	Mar 19, 2019
Grant date	Mar 19, 2019

Abstract

Method for solving the decomposition-coordination calculation based on Block Bordered Diagonal Form (BBDF) model by using data center. During the solving process, partitioning the electric power system network by using the existing network partitioning method to achieve the grid partition, and setting the parameters of virtual memories firstly, thus to establish the bin-packing model with the priority of energy efficiency; and then, setting each calculating step of the decomposition-coordination calculation based on BBDF as a task. Through the manners that servers host VMs and VMs map tasks, the decomposition-coordination algorithm can be executed in data center, and the running time and energy consumption of data center can be calculated. The calculating time of decomposition-coordination algorithm is shortened and the energy consumption in data center. Moreover, with the increase of scale and the complexity of the electric power system, the advantages of the method using data center presented by the present invention are becoming much more obvious.

First claim

What is claimed is:

1. A method for solving the decomposition-coordination calculation based on BBDF by using data center, which comprises the following steps:
(1) partitioning the electric power system network by using the existing network partitioning method to achieve the grid partition;
(2) setting each calculating step of the decomposition-coordination calculation based on BBDF as a task, and counting the total number of tasks of calculating process;
(3) acquiring the values of MIPS (Million Instructions Per Second) and memory in each PMs of data center, setting the number of VMs, and setting the values of MIPS and memory in each Virtual Machine (VM);
(4) calculating the total energy consumption of Information Technology (IT) equipments in data center by the following formula:

E
total

=

∑

i
=
1

N

⁢

⁢

E
server

+

∑

j
=
1

T

⁢

⁢

E
switch

(
1
)

wherein, E server represents the energy consumption of single server in the data center, E switch represents the energy consumption of single switch in the data center, N represents the number of servers in the data center, T represents the number of switches in the data center;
the energy consumption model of single server is expressed as follows:

E
server

=

P
baseline

⁢

t
max

+

P
VM

⁢

∑

k
=
1

M

⁢

⁢

t
k

(
2
)

wherein, Pbaseline represents the power consumption of the no-load running server, tmax represents the whole running time of servers, PVM represents the power consumption of a virtual machines in servers, tkrepresents the running time of the virtual machine k in servers, M represents the amount of virtual machines in servers;
the energy consumption model of single switch is expressed as follows:
E switch =P switch t max   (3)

wherein, Pswitch represents the running power consumption of switches, tmax represents the whole running time of switches;
(5) using bin-packing model to describe the server energy consumption model, and achieving the minimum energy consumption model of servers;

the set of virtual machines is: V={V 1 , V 2 , . . . , V Q |V j =(R j MIPS , R j MEM )}, wherein R j MIPS and R j MEM represent the values of MIPS and memory respectively for one virtual machine, the capacity value of single server is C i =(C i MIPS ,C i MEM ), wherein C i MIPS and C i MEM represent the values of MIPS and memory respectively in a single server; the total servers' energy consumption model of the data center is expressed as follows:

min
⁢

⁢

∑

i
=
1

⁢

⁢

E
server

⁢

H
i

(
4
)

the constraints of the total servers' energy consumption model are as follows:
∑ j = 1 M ⁢ ⁢ R j MIPS · X j , i ≤ C i MIPS , ∀ i ∈ { 1 , 2 , 3 ⁢ ⁢ … ⁢ } ( 5 ) ∑ j = 1 M ⁢ ⁢ R j MEM · X j , i ≤ C i MEM , ∀ i ∈ { 1 , 2 , 3 ⁢ ⁢ … ⁢ } ( 6 ) X j,i =0 or 1  (7)

H i =0 or 1  (8)

wherein, X j,i =1 represents the virtual machine j is packed in the server i, X j,i =0 represents the virtual machine j is not in the server I; if the server i is used, H i =1, otherwise, H i =0;
(6) adopting best-fit algorithm or descending order best-fit algorithm to calculate the total energy consumption of data center servers;
(7) packing the VM mentioned in step (3) into the servers;
(8) packing every task mentioned in step (2

Assignees

• Univ Tianjin

Inventors

• Yang Ting
• Xiang Wenping
• Feng Yingmin
• Pen Haibo
• Xu Mingyu
• You Jinkuo
• Wang Hongtao

Classifications

• G06F2009/45583

  Memory management, e.g. access or allocation · CPC title

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F11/36

  Prevention of errors by analysis, debugging or testing of software · CPC title