CN105389448B - A kind of graph model building method for protecting the assessment of energy degree for computer cluster - Google Patents
A kind of graph model building method for protecting the assessment of energy degree for computer cluster Download PDFInfo
- Publication number
- CN105389448B CN105389448B CN201510975124.6A CN201510975124A CN105389448B CN 105389448 B CN105389448 B CN 105389448B CN 201510975124 A CN201510975124 A CN 201510975124A CN 105389448 B CN105389448 B CN 105389448B
- Authority
- CN
- China
- Prior art keywords
- node
- graph model
- energy degree
- computer
- assessment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/02—CAD in a network environment, e.g. collaborative CAD or distributed simulation
Abstract
The present invention relates to a kind of graph model building method for protecting the assessment of energy degree for computer cluster, its step with the following method: step 1: it is configured to computer cluster and protects the tree-model that energy degree is assessed；Step 2: graph model conversion is carried out using the method for merge node；Step 3: graph model conversion is carried out using the method for useless knot removal；Step 4: graph model conversion is carried out using the method that redundant node is deleted；It is insufficient that it solves system mode enumeration methodology performance, it is difficult to carry out the problem of large-scale computer cluster protects the assessment of energy degree, it has many advantages, such as that making to protect the assessment of energy degree faster more saves resource.
Description
Technical field
The present invention relates to a kind of graph model building methods that the assessment of energy degree is protected for computer cluster.
Background technique
Computer cluster is the aggregate of one group of independent computer, and intercomputer is connected by high performance internet
It connects, each computer, which can cooperate and show as a computing resource that is single, concentrating, provides service to the network user.Meter
Calculation machine cluster be it is a kind of it is cheap, be easy to construct and the parallel machine architecture with preferable scalability.Cluster it is each
It is independent from each other between computer, and there is different performance numbers.For example, the different IBM in a computer cluster
The mostly specific performance number of computer, HP computer and thinking computer often has very big difference.And entire computer cluster
Performance be all computer performance values for forming the clusters cumulative (summarizing).It is each in computer cluster in the process of running
A computer will appear random fault, and when a computer breaks down, which, which contributes the performance number of entire cluster, is
0, and when a computer does not break down and works normally, which is the computer to the performance number contribution of entire cluster
Performance number.
The guarantor's energy degree for assessing computer cluster is exactly to assess computer set under conditions of random fault can occur for computer
The performance number of group is in the probability in some defined section.This section, which also becomes, protects energy degree assessment section.It exists asks as follows
Topic: existing computer cluster protects the assessment of energy degree and is normally based on the method that system mode is enumerated, when the computer that cluster includes
Quantity N increases, system mode quantity 2 possessed by clusterNJust sharply increase.So that enumeration methodology be only applicable to it is small-sized
Computer cluster, it is more difficult to the assessment of large-scale computer cluster, it is difficult to obtain accurate assessed value.
Summary of the invention
In view of the defects and deficiencies of the prior art, the present invention intends to provide a kind of structure is simple, design rationally, make
With a kind of convenient graph model building method for protecting the assessment of energy degree for computer cluster, it solves system mode enumeration methodology
Energy is insufficient, it is difficult to carry out the problem of large-scale computer cluster protects the assessment of energy degree, it, which has, saves guarantor's energy degree assessment faster more
The advantages that resource.
To achieve the above object, the technical solution adopted by the present invention is that:
A kind of graph model building method for protecting the assessment of energy degree for computer cluster of the present invention, it is using such as lower section
Method step:
Step 1: it is configured to the tree-model that computer cluster protects the assessment of energy degree；Wherein, guarantor's energy degree of computer cluster is just
It is to assess the performance number of computer cluster under conditions of random fault can occur for computer to be in the general of section as defined in some
Rate, this section, which also becomes, protects energy degree assessment section；
Step 2: being directed to step 1 tree-model obtained, carries out graph model conversion using the method for merge node；
Step 3: being directed to step 2 graph model obtained, carries out graph model conversion using the method for useless knot removal；
Step 4: being directed to step 3 graph model obtained, carries out graph model conversion using the method that redundant node is deleted.
Further, in the step 1, for ' normal ' possessed by computer each in computer cluster and ' event
Hinder ' two state and each computer possessed performance number under ' normal ' state, establishes corresponding state space tree mould
Type, and defined according to the evaluation interval for protecting energy degree, the leaf node of tree-model is set.
Further, in the step 2, to each node layer in step 1 tree-model obtained, had according to node
Some summarizes performance number and carries out isomorphism judgement, when summarizing performance number there are two nodes are having the same in one layer, then merges this
Two nodes
Further, in the step 3, to each node in step 2 graph model obtained, had according to node
The performance number that summarizes judged, if currently summarize performance number exceeded protect energy degree the evaluation interval upper limit, the node
It is that useless node directly can replace the node with leaf node ' 0 '；Summarize if currently summarizing performance number and adding remaining maximum
Performance number is still not able to be greater than the evaluation interval lower limit for protecting energy degree, then the node is that useless node can directly use leaf node
' 0 ' replaces the section.
Further, in the step 4, to each node in step 3 graph model obtained, if two of node
Branch is directed toward identical child node, then the node is that redundant node can use child node and replace the node.
After adopting the above structure, the invention has the following beneficial effects: it is of the present invention a kind of for computer cluster guarantor's energy degree
The graph model building method of assessment, it is insufficient that it solves system mode enumeration methodology performance, it is difficult to carry out large-scale computer cluster
The problem of protecting the assessment of energy degree, it has many advantages, such as that making to protect the assessment of energy degree faster more saves resource.
Detailed description of the invention
Fig. 1 is that computer cluster example guarantor's energy degree evaluation status enumerates model；
Fig. 2 is that computer cluster example protects energy degree assessment tree-model；
Fig. 3 is guarantor's energy degree assessment graph model after the method processing by merge node；
Fig. 4 is that guarantor's energy degree after the method processing by useless knot removal assesses graph model；
Fig. 5 is that guarantor's energy degree after the method processing deleted by redundant node assesses graph model；
Specific embodiment
The present invention will be further described below with reference to the drawings.
As shown in Figure 1, a kind of graph model building method that the assessment of energy degree is protected for computer cluster of the present invention, it
Step with the following method:
Step 1: it is configured to the tree-model that computer cluster protects the assessment of energy degree；Wherein, guarantor's energy degree of computer cluster is just
It is to assess the performance number of computer cluster under conditions of random fault can occur for computer to be in the general of section as defined in some
Rate, this section, which also becomes, protects energy degree assessment section；
Step 2: being directed to step 1 tree-model obtained, carries out graph model conversion using the method for merge node；
Step 3: being directed to step 2 graph model obtained, carries out graph model conversion using the method for useless knot removal；
Step 4: being directed to step 3 graph model obtained, carries out graph model conversion using the method that redundant node is deleted.
Preferably as one kind of the invention, in the step 1, for possessed by computer each in computer cluster
' normal ' and ' failure ' two states and each computer possessed performance number under ' normal ' state are established corresponding
State space tree-model, and defined according to the evaluation interval for protecting energy degree, the leaf node of tree-model is set.
Preferably as one kind of the invention, in the step 2, to each node layer in step 1 tree-model obtained,
Summarize performance number according to possessed by node and carry out isomorphism judgement, when summarizing performance there are two nodes are having the same in one layer
Value, then merge the two nodes
Preferably as one kind of the invention, in the step 3, to each node in step 2 graph model obtained, root
Judged according to performance number is summarized possessed by node, has been exceeded on the evaluation interval for protecting energy degree if currently summarizing performance number
Limit, then the node is that useless node directly can replace the node with leaf node ' 0 '；If currently summarizing performance number plus surplus
Remaining maximum summarizes performance number and is still not able to be greater than the evaluation interval lower limit for protecting energy degree, then the node is that useless node can be direct
The section is replaced with leaf node ' 0 '.
Preferably as one kind of the invention, in the step 4, to each node in step 3 graph model obtained, such as
The Liang Ge branch of fruit node is directed toward identical child node, then the node is that redundant node can use child node and replace the node.
Working principle of the present invention is as follows:
Step 1: it is configured to the tree-model that computer cluster protects the assessment of energy degree；
For ' normal ' and ' failure ' two states and each calculating possessed by computer each in computer cluster
Machine possessed performance number under ' normal ' state is established corresponding tree-model, and is defined according to the evaluation interval for protecting energy degree, if
Set the leaf node of tree-model.
By taking computer cluster A as an example.
Computer cluster A includes 4 computers.The performance number of computer 1 is 1, and the performance number of computer 2 is 1, computer
3 performance number is 0.5, and the performance number of computer 4 is 0.5.The evaluation interval for protecting energy degree is [1,1.5], i.e. clustering performance value
Lower limit is that 1 upper limit is 1.5.
State as shown in Figure 1 can be constructed directly for the computer cluster and enumerate model.Each of tree-model is non-
Bottom layer node all corresponds to a computer, for example top mode corresponds to computer 1, and the second node layer corresponds to 2. tree-model of computer
In each node have 2 branches, branch 1 indicates that computer be in normal condition, and the expression computer of branch 0 is in failure shape
State.Each node in tree-model has one currently to summarize performance number.Illustrate currently to summarize performance number below by example
Calculation method.For example, one of the 4th layer marked in Fig. 1 belongs to the node of computer 4, needed from top mode to the node
The condition to be met is: computer 1 is failure, and computer 2 and computer 3 are normally, so the computer 4 of this mark
The performance number that currently summarizes of node is 2 performance number of computer plus 3 performance number of computer, i.e. 1+0.5=1.5. in another example, Fig. 1
One bottom layer node of middle marked layer 5 needs the condition met to be: computer 2 and meter from top mode to the node
Calculation machine 3 is failure, and computer 1 and computer 4 are normal, so the bottom layer node of this mark currently summarizes performance number
4 performance number of computer, i.e. 1+0.5=1.5 are added for 1 performance number of computer.
Enumerating model for the state in Fig. 1 can define according to the evaluation interval for protecting energy degree, and bottom layer node is converted into
For leaf node.The rule of conversion is:, should if the performance number that currently summarizes of bottom layer node is in the evaluation interval for protecting energy degree
Bottom layer node is converted to leaf node 1,；If the performance number that currently summarizes of bottom layer node is not at the evaluation interval for protecting energy degree,
Then the bottom layer node is converted to leaf node 0.It finally obtains computer cluster example shown in Fig. 2 and protects energy degree assessment tree-model.
Step 2: the method using merge node carries out graph model conversion；
To each node layer in step 1 tree-model obtained, performance number progress is currently summarized according to possessed by node
Isomorphism judgement then merges the two nodes when currently summarizing performance number there are two nodes are having the same in one layer.
By taking computer cluster A as an example.
In Fig. 2 two of third layer belong to computer 3 node it is having the same currently summarize performance number (being all 1), then
The two nodes can be merged, obtain guarantor's energy degree assessment graph model after the method processing shown in Fig. 3 by merge node.
Step 3: the method using useless knot removal carries out graph model conversion；
To each node in step 2 graph model obtained, currently summarizes performance number according to possessed by node and sentenced
Disconnected, if currently summarizing performance number has exceeded the evaluation interval upper limit for protecting energy degree, which is that useless node can be direct
The node is replaced with leaf node ' 0 '；It is still not able to greatly if currently summarizing performance number and summarizing performance number plus remaining maximum
The evaluation interval lower limit of Yu Baoneng degree, then the node is that useless node directly can replace the node with leaf node ' 0 '.
By taking computer cluster A as an example.
One of third layer belongs to the node of computer 3 in Fig. 3, currently summarizes performance number and has exceeded commenting for guarantor's energy degree
The valence section upper limit, so the node is that useless node directly can replace the node with leaf node ' 0 '.4th layer of a category
In the node of computer 4, currently summarizes performance number 0 and add remaining maximum and summarize performance number 0.5 and be still not able to be greater than and protect energy
The evaluation interval lower limit 1 of degree, so the node is that useless node directly can replace the node with leaf node ' 0 '.Finally obtain
Obtain guarantor's energy degree assessment graph model after the method processing shown in Fig. 4 by useless node.
Step 4: carrying out graph model conversion using the method that redundant node is deleted.
To each node in step 3 graph model obtained, if 2 branches of node are directed toward identical child node,
The node is that redundant node can use child node and replace the node.
By taking computer cluster A as an example.
One of the 4th layer belongs to the node of computer 4 in Fig. 3, and 2 branches of the node are directed toward identical child node, i.e.,
Leaf node 1, so the node is that redundant node can use child node (leaf node 1) and replace the node.It obtains shown in Fig. 5
By useless node method processing after guarantor's energy degree assessment graph model.
Implementation result of the invention: finding out from the comparison of Fig. 1 and Fig. 5, and Fig. 1 is using locating for traditional state enumeration methodology
The model of reason, Fig. 5 are using model handled by graph model building method provided by the invention.Model of the invention is much smaller than biography
The model for enumeration methodology of uniting, so that protecting the assessment of energy degree faster more saves resource.
A kind of graph model building method for protecting the assessment of energy degree for computer cluster of the present invention, it solves system shape
State enumeration methodology performance is insufficient, it is difficult to carry out the problem of large-scale computer cluster protects the assessment of energy degree, it, which has, comments guarantor's energy degree
Estimate the advantages that faster more saving resource
The above description is only a preferred embodiment of the present invention, thus it is all according to the configuration described in the scope of the patent application of the present invention,
The equivalent change or modification that feature and principle are done, is included in the scope of the patent application of the present invention.
Claims (4)
1. a kind of graph model building method for protecting the assessment of energy degree for computer cluster, it is characterised in that: it is with the following method
Step:
Step 1: it is configured to the tree-model that computer cluster protects the assessment of energy degree；Wherein, guarantor's energy degree of computer cluster is exactly to comment
The performance number for estimating computer cluster under conditions of random fault can occur for computer is in the probability in section as defined in some, this
A section, which also becomes, protects energy degree assessment section；
Step 2: being directed to step 1 tree-model obtained, carries out graph model conversion using the method for merge node；To step 1
Each node layer in tree-model obtained summarizes performance number according to possessed by node and carries out isomorphism judgement, deposits when in one layer
Summarize performance number two nodes are having the same, then merges the two nodes；
Step 3: being directed to step 2 graph model obtained, carries out graph model conversion using the method for useless knot removal；
Step 4: being directed to step 3 graph model obtained, carries out graph model conversion using the method that redundant node is deleted.
2. a kind of graph model building method for protecting the assessment of energy degree for computer cluster according to claim 1, feature
It is: in the step 1, for ' normal ' and ' failure ' two states possessed by computer each in computer cluster, with
And each computer possessed performance number under ' normal ' state, corresponding state space tree-model is established, and according to guarantor's energy
The evaluation interval of degree defines, and the leaf node of tree-model is arranged.
3. a kind of graph model building method for protecting the assessment of energy degree for computer cluster according to claim 1, feature
It is: in the step 3, to each node in step 2 graph model obtained, performance number is summarized according to possessed by node
Judged, if currently summarizing performance number has exceeded the evaluation interval upper limit for protecting energy degree, which is that useless node can
Directly to replace the node with leaf node ' 0 '；Summarize performance number still not if currently summarizing performance number and adding remaining maximum
The evaluation interval lower limit for protecting energy degree can be greater than, then the node is that useless node directly can replace the section with leaf node ' 0 '.
4. a kind of graph model building method for protecting the assessment of energy degree for computer cluster according to claim 1, feature
It is: in the step 4, to each node in step 3 graph model obtained, if the Liang Ge branch direction of node is identical
Child node, then the node is that redundant node can use child node and replace the node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510975124.6A CN105389448B (en) | 2015-12-22 | 2015-12-22 | A kind of graph model building method for protecting the assessment of energy degree for computer cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510975124.6A CN105389448B (en) | 2015-12-22 | 2015-12-22 | A kind of graph model building method for protecting the assessment of energy degree for computer cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105389448A CN105389448A (en) | 2016-03-09 |
| CN105389448B true CN105389448B (en) | 2019-07-23 |
Family
ID=55421732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510975124.6A Active CN105389448B (en) | 2015-12-22 | 2015-12-22 | A kind of graph model building method for protecting the assessment of energy degree for computer cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105389448B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1601510A (en) * | 2003-03-06 | 2005-03-30 | 微软公司 | Architecture for distributed computing system and automated design, deployment, and management of distributed applications |
| CN104158840A (en) * | 2014-07-09 | 2014-11-19 | 东北大学 | Method for calculating node similarity of chart in distributing manner |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8589855B1 (en) * | 2012-05-30 | 2013-11-19 | International Business Machines Corporation | Machine-learning based datapath extraction |
-
2015
- 2015-12-22 CN CN201510975124.6A patent/CN105389448B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1601510A (en) * | 2003-03-06 | 2005-03-30 | 微软公司 | Architecture for distributed computing system and automated design, deployment, and management of distributed applications |
| CN104158840A (en) * | 2014-07-09 | 2014-11-19 | 东北大学 | Method for calculating node similarity of chart in distributing manner |
Non-Patent Citations (2)
| Title |
|---|
| A Multiple-Valued Decision Diagram Based Method for Efficient Reliability Analysis of Non-Repairable Phased-Mission Systems;Mo Y 等;《IEEE Transactions on Reliability》;20140331;第63卷(第1期);第320-330页 |
| 一种基于移动服务器端的树图建模方法;刘炜 等;《计算机科学》;20110430;第38卷(第4期);第55-60页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105389448A (en) | 2016-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN207301773U (en) | A kind of numerical control machine tool monitoring system based on Internet of Things | |
| CN103502942A (en) | Virtual machine administration device, virtual machine administration method, and program thereof | |
| CN102279905B (en) | Method for rapidly reducing data streams during power grid fault diagnosis | |
| CN108733464A (en) | A kind of method and device of the scheduling scheme of determining calculating task | |
| CN105243004A (en) | Failure resource detection method and apparatus | |
| CN101883003A (en) | Mandatory running method of computer operating system | |
| CN104133727A (en) | Load distribution method based on real-time resources | |
| CN105938502A (en) | Validation method for design of system security of AltaRica model | |
| CN102521265A (en) | Dynamic consistency control method in massive data management | |
| CN107104820B (en) | Dynamic capacity-expansion daily operation and maintenance method based on F5 server node | |
| CN102811152B (en) | Method for realizing real-time transaction and data exchange of multiple main bus network communication | |
| JP2015109074A (en) | Automated generation and dynamic update of rules | |
| CN111418187A (en) | Scalable statistics and analysis mechanism in cloud networks | |
| CN102761448A (en) | Cluster monitoring and early warning method | |
| CN105389448B (en) | A kind of graph model building method for protecting the assessment of energy degree for computer cluster | |
| CN110046064B (en) | Cloud server disaster tolerance implementation method based on fault drift | |
| CN104572467A (en) | Method for testing performance of virtual platform database of server | |
| Guo et al. | Cyber-physical failure detection system: Survey and implementation | |
| CN108280008A (en) | One kind being directed to Hadoop cluster abnormal nodes method of real-time | |
| CN108712499A (en) | Based on the system being managed to cloud computing service | |
| CN108282360A (en) | A kind of fault detection method of shot and long term prediction fusion | |
| CN105721199A (en) | Real-time cloud service bottleneck detection method based on kernel density estimation and fuzzy inference system | |
| CN106357427A (en) | Fault computer processing method and processing system | |
| CN103327093B (en) | The control method of cloud computing system | |
| CN102223250A (en) | Method for switching industrial redundancy server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190621 Address after: 362021 No. 269 North China Road, Fengze District, Quanzhou City, Fujian Province Applicant after: Huaqiao University Address before: No. 688 Yingbin Road, Jinhua, Zhejiang Province, Zhejiang Applicant before: Zhejiang Normal University |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |