CN109688758A - For determining method, apparatus, electronic equipment and the medium of data center's failure - Google Patents

For determining method, apparatus, electronic equipment and the medium of data center's failure Download PDF

Info

Publication number
CN109688758A
CN109688758A CN201910110034.9A CN201910110034A CN109688758A CN 109688758 A CN109688758 A CN 109688758A CN 201910110034 A CN201910110034 A CN 201910110034A CN 109688758 A CN109688758 A CN 109688758A
Authority
CN
China
Prior art keywords
node
layer
node device
equipment
data center
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.)
Granted
Application number
CN201910110034.9A
Other languages
Chinese (zh)
Other versions
CN109688758B (en
Inventor
张彻
张实唯
李伟超
汪漪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest University of Science and Technology
Original Assignee
Southwest University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southwest University of Science and Technology filed Critical Southwest University of Science and Technology
Priority to CN201910110034.9A priority Critical patent/CN109688758B/en
Publication of CN109688758A publication Critical patent/CN109688758A/en
Application granted granted Critical
Publication of CN109688758B publication Critical patent/CN109688758B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K7/00Constructional details common to different types of electric apparatus
    • H05K7/14Mounting supporting structure in casing or on frame or rack
    • H05K7/1485Servers; Data center rooms, e.g. 19-inch computer racks
    • H05K7/1498Resource management, Optimisation arrangements, e.g. configuration, identification, tracking, physical location

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

Method, apparatus, electronic equipment and the medium that the embodiment of the invention discloses a kind of for determining data center's failure, wherein this method comprises: node device each in physical map is divided at least one node layer;In each node layer, according to the similarity between any two node device and the connection relationship of other node layer interior joint equipment, the node device in each node layer is divided into the node group of preset quantity;According to the connection relationship of node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint of data center, parameter tags are carried out to the node device in each node group;According to the connection relationship between node role each in devising and drawing the blueprint, intermediary matrix figure is created based on each node device after parameter tags;Malfunctioning node equipment and its node role are determined based on physical map and intermediary matrix figure.The embodiment of the present invention realizes the malfunctioning node for efficiently determining data center, and improves the accuracy that malfunctioning node determines.

Description

For determining method, apparatus, electronic equipment and the medium of data center's failure
Technical field
The present embodiments relate to technical field of the computer network more particularly to a kind of for determining data center's failure Method, apparatus, electronic equipment and medium.
Background technique
In order to provide the user with high bandwidth low latency, stabilization and reliable network service, many application programs newly developed It is deployed in cloud environment, this has pushed storage, transmission and has calculated service from computer client to data center The transformation of (Datacenter, abbreviation DC).In order to meet more and more demands for services in single computer client, such as network The scale of search, video-see, mobile Internet and cloud storage etc., service terminal exponentially increases.Construct reasonable data Central site network topology and routing, to realize data center's Quick Extended and make full use of data center network performance, become calculating The important research direction of machine network field.
In the prior art according to the computer room blueprint for the data matrix form being pre-designed, it is based on Graph Theory, and record Equipment and position corresponding relationship, construct data center.But the building process is comparatively laborious, and records each and set The standby corresponding relationship with position is a huge job, and this record may be easy to malfunction, and data center is caused to construct Error.
What existing Fault Locating Method used when generalling use based on data center's building devise and draw the blueprint and from building up The physical map that data center collects, the public subgraph of maximum for determining to devise and draw the blueprint between physical map, the then event of data center The i.e. corresponding difference section to devise and draw the blueprint between physical map in barrier part.Simultaneously, it is assumed that administrator oneself can have found and repair The failure for thering is node degree to change, and the failure constant for node degree, first devise and draw the blueprint in physical map it is preliminary really The corresponding node of possibility for determining preset quantity, then gradually extends subgraph as anchor point using these nodes, and judges from setting The subgraph that expands of anchor point in meter blueprint and physical map whether isomorphism, when judging two subgraph tripe systems of extension, then Determine that malfunctioning node is present in current subgraph.
The accuracy of above-mentioned Fault Locating Method depends on the mistake in the anchor point quantity and network for detection of selection The quantity of wiring.Also, it is not only very time-consuming in the case where no correctly selection anchor point, and accuracy rate is lower.Consider to It is real malfunctioning node that result out is different, which to establish a capital, maintenance personnel just need to first check in the report of fault location result about The long list of possible breakdown node, the then real malfunctioning node of manually identifying.
Summary of the invention
The embodiment of the present invention provides a kind of method, apparatus for determining data center's failure, electronic equipment and medium, with It realizes the malfunctioning node for efficiently determining data center, and improves the accuracy that malfunctioning node determines.
In a first aspect, the embodiment of the invention provides a kind of methods for determining data center's failure, this method comprises:
Node device each in the physical map of data center is divided at least one node layer according to device type;
In each node layer, according to the connection relationship of any two node device and other node layer interior joint equipment it Between similarity, the node device in each node layer is divided into the node group of preset quantity;
According to the marking convention of the interior joint role that devises and draw the blueprint of the data center and the intermediate node of each node layer The connection relationship of equipment carries out parameter tags to the node device in each node group, node device after making parameter tags with It is described devise and draw the blueprint in node role correspond;
According to the connection relationship between node role each in described devise and draw the blueprint, based on each node after parameter tags Equipment create intermediary matrix figure, wherein position of each node device in the intermediary matrix figure in the physical map Position it is identical;
Determine the target difference matrix element in the physical map and the intermediary matrix figure, and according to the target difference Matrix element determines malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment.
Second aspect, the embodiment of the invention also provides a kind of for determining the device of data center's failure, the device packet It includes:
Node layer division module, for node device each in the physical map by data center according to device type be divided into A few node layer;
Node group division module is used in each node layer, according in any two node device and other node layers Node device in each node layer is divided into the node of preset quantity by the similarity between the connection relationship of node device Group;
Parameter tags module, for the marking convention of the interior joint role that devises and draw the blueprint according to the data center, and The connection relationship of node device between each node layer carries out parameter tags to the node device in each node group, makes parameter Node device after label is corresponded with the node role in described devise and draw the blueprint;
Intermediary matrix figure creation module, for according to the connection relationship between node role each in described devise and draw the blueprint, Intermediary matrix figure is created based on each node device after parameter tags, wherein each node device is in the intermediary matrix figure In position it is identical as the position in the physical map;
Fault determination module, for determining the target difference matrix element in the physical map and the intermediary matrix figure, And malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment are determined according to the target difference matrix element.
The third aspect, the embodiment of the invention also provides a kind of electronic equipment, comprising:
One or more processors;
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing Device realizes the method for determining data center's failure as described in any embodiment of the present invention.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer Program realizes the side for being used to determine data center's failure as described in any embodiment of the present invention when the program is executed by processor Method.
The embodiment of the present invention is layered and is grouped by node device each in the physical map to data center, according to number It is right according to the connection relationship of node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint at center Node device in each node group carries out parameter tags, and closes according to the connection between node role each in devising and drawing the blueprint System creates intermediary matrix figure based on each node device after parameter tags, and the comparison based on physical map and intermediary matrix figure is true Determine malfunctioning node equipment and its node role, it is more lower than relatively time-consuming and accuracy to solve malfunctioning node determination in existing method The problem of, the malfunctioning node for efficiently determining data center is realized, and improve the accuracy that malfunctioning node determines.
Detailed description of the invention
Fig. 1 is the flow chart for the method for determining data center's failure that the embodiment of the present invention one provides;
Fig. 2 is provided by Embodiment 2 of the present invention for determining the flow chart of the method for data center's failure;
Fig. 3 is the flow chart for the method for determining data center's failure that the embodiment of the present invention three provides;
Fig. 4 a is a kind of schematic diagram for intermediate blueprint that the embodiment of the present invention three provides;
Fig. 4 b is the schematic diagram for the intermediate blueprint of another kind that the embodiment of the present invention three provides;
Fig. 4 c is a kind of schematic diagram for intermediary matrix figure that the embodiment of the present invention three provides;
Fig. 5 is the structural schematic diagram for the device for determining data center's failure that the embodiment of the present invention four provides;
Fig. 6 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention five provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart for the method for determining data center's failure that the embodiment of the present invention one provides, the present embodiment It is applicable to the case where physical map according to data center determines the malfunctioning node of connection error in data center, this method can be with By executing for determining the device of data center's failure, which can be realized by the way of software and/or hardware, and can It integrates on an electronic device.
As shown in Figure 1, the method provided in this embodiment for determining data center's failure may include:
S110, node device each in the physical map of data center is divided at least one node layer according to device type.
Physical map is used to record the connection relationship in the data center built up between each node device, and detection The basis of connecting fault between node device existing for data center.After the physical map creation demand of data center is triggered, Electronic equipment for executing this embodiment scheme can be unified to send inquiry instruction to each node device of data center;Respectively A node device in response to the inquiry instruction, and by respective port connection relationship, respective physical address (MAC Address) and The information such as respective logical ip are sent to the electronic equipment;The electronic equipment is connected based on the port of received each node device The information such as relationship, MAC Address and logical ip obtain the physical map of data center by data processing.Wherein, pre- in physical map First certain specified a line and a certain column, such as the first row and first row, can respectively indicate the MAC Address of each node device, or The logical ip of each node device of person, for realizing the differentiation in physical map to each node device;Node device is removed to occupy Row and column except matrix element just indicate the connection relationship between corresponding two node devices, for example, if two nodes There are connection relationships then to be indicated using number 1 between equipment, uses number if connection relationship is not present between two node devices Word 0 indicates.On the basis of may be implemented to distinguish indicates between node device with the presence or absence of connection relationship, the present embodiment is to used Character style be not especially limited.
According to the difference of deployed with devices in different data centers, the device type for including in physical map is also not quite similar. Different types of node device can have differences, for example, connection degree of the different types of node device in physical map may Difference, the connection degree refer to the quantity for other node devices connecting with each node device;Belong to the number of same layer node device Amount is that the quantity of same type of node device may be different, therefore, can be based on the node device of connection degree and same type The factors such as quantity are layered the node device in physical map, and each layer corresponds to a device type, to realize node The preliminary classification of equipment.
Illustratively, the connection relationship net between data center's interior joint equipment is in upper layer such as inverted pyramid shape Node device connection degree be greater than in lower layer node device connection degree, can be by each in Statistical Physics figure The node device for being less than current degree threshold value is determined as same layer node device by the connection degree of node device;When a node After layer determines, being adapted to property adjusts degree threshold value, then the node device less than the degree threshold value after adjustment is determined as Same layer node device repeats aforesaid operations, until each node device in physical map is all divided to corresponding node layer. The degree threshold value in operating process can carry out adaptability setting according to the deployed with devices situation of data center each time.
When the connection degree of different types of node device there is a situation where it is identical, such as based on fat tree topology network configuration (FatTree) in the data center constructed, other than the node device of type of server, the node device of all switch types Connection degree it is all the same, the mode of the connection degree that is based purely on statistics and degree threshold value comparison can not achieve all nodes Equipment layering just needs that other operations is combined to realize layering.Optionally, node device each in the physical map of data center is pressed At least one node layer is divided into according to device type, comprising: determines the Connected degree of each node device in the physical map of data center Number will connect node device of the degree less than current degree threshold value and be determined as same layer node device, i.e., will connect in present physical figure It connects the smallest node device of degree and determines and be divided into a node layer, and by same layer node device and its connection relationship It is removed from physical map, obtains current SWITCH figure;Repeat the behaviour that same layer node device is determined in current SWITCH figure Make the operation with removal same layer node device and its connection relationship, until each node device is divided into phase in physical map The node layer answered.
According to the connection relationship between each node device, when determine first in physical map it is minimum by connection degree Node device composition node layer after, the node layer equipment and its connection relationship are removed from physical map, with the layer There are the connection degrees of the node device of connection relationship to reduce, and therefore, will occur again in remaining node device One group of the smallest node device of connection degree, thus the determination by repeating same layer node device and removal operate, it can be with Each node device in physical map is divided to corresponding node layer.Specifically, sequence is executed according to repetitive operation, according to The node layer of secondary determination is server node layer, edge switch node layer, aggregation switch node layer and core switch section The connection degree of point layer, i.e. server node layer interior joint equipment is minimum.Also, after determining edge switch node layer, It there are the remaining node device of connection relationship is aggregation switch node layer with the node device in edge switch node layer In node device, further with aggregation switch node layer interior joint equipment there are the remaining node devices of connection relationship just It is the node device of core switch node layer.It should be noted that the specific divided node layer of physical map interior joint equipment 5 layers can also be less than, it can including server node layer, edge switch node layer, aggregation switch node layer and core One layer in telephone net node layer or multilayer, this is related with the specifically deployment of data center, but is in the section of data center The connection degree of the point device connection undermost node device of network is still the smallest.
S120, in each node layer, according to the connection of any two node device and other node layer interior joint equipment Node device in each node layer is divided into the node group of preset quantity by the similarity between relationship.
There are an implementation premise in the present embodiment, i.e., each layer of node device quantity is even number, therefore, Ke Yishi The grouping two-by-two of existing each node layer interior joint equipment.The node device for belonging to the same node group is closed with similar connection System.Due to the malfunctioning node equipment in physical map there are connection error, do not ensure that each node in each node layer Connection relationship has the similitude of same degree between node device in group.Follow two nodes that connection relationship is most like Equipment is divided into the principle of same node group, last remaining two node devices, no matter their connection relationship similarity degree How, it is divided into a node group for the time being.It is in order to subsequent convenient for according to data by the way that same layer node device to be grouped Center devise and draw the blueprint in physical map each node device carry out parameter tags, thus realize physical map interior joint equipment with Devise and draw the blueprint the one-to-one correspondence of interior joint role.Therefore, even if there are the node devices of connection relationship dissmilarity in grouping process It is divided into same node group, the process of subsequent determining malfunctioning node equipment can also be determined again, will not influence failure The accuracy of positioning.
S130, according to data center the interior joint role that devises and draw the blueprint marking convention and each node layer segmentum intercalaris The connection relationship of point device carries out parameter tags to the node device in each node group, the node device after making parameter tags It is corresponded with the node role in devising and drawing the blueprint.
Devise and draw the blueprint be according to the role of node device predetermined, the building demand at combined data center and design , the building for guide data center, wherein node role predetermined is identical as the affiliated type of the role for representing At least one similar node device, node role's connection relationship reflects between node device corresponding with node role just True connection relationship.It include the connection relationship between the node device planned in advance in the building demand of data center.It is designing In blueprint, according to preset angle of rotation of joint color marker rule, each node role is marked using corresponding character parameters, node The flag parameters of role can also react the connection relationship of node role between different role layer.According to the classification of node role, Each node role under same type can correspond to any node equipment identical with character types, such as belong to interchanger class Each node role of type, can be to any interchanger in the node device that should belong to switch type.Therefore, it is designing In blueprint, there is no embody the node device of the corresponding same type of each node role and uniqueization.
The angle of rotation of joint color marker rule to devise and draw the blueprint define how to node role carry out parameter tags, at this point, according to The angle of rotation of joint color marker rule, carries out parameter tags to the node device of node layer each in physical map.Specifically, when determining After the corresponding node layer of physical map interior joint equipment, the angle of rotation of joint chromatograph corresponded in devising and drawing the blueprint is just determined, according to node The parameter tags rule of each node role carries out parameter tags to the node device in corresponding physical map in role's layer.
S140, according to the connection relationship between node role each in devising and drawing the blueprint, based on each section after parameter tags Point device creates intermediary matrix figure, wherein position of each node device in intermediary matrix figure and the position in physical map It is identical.
The connection relationship between each node role that devises and draw the blueprint is that corresponding node device each into physical map is due just True connection relationship, therefore, when connection error occurs in the connection relationship between data center's building process interior joint equipment, design is blue Just there is inconsistent phenomenon in the connection relationship reflected in figure and physical map.Utilize correctly connecting for the interior joint role that devises and draw the blueprint Relationship is connect, intermediary matrix figure is obtained, determines that there are the node devices of connection relationship mistake in physical map by intermediary matrix figure, i.e., Malfunctioning node equipment.
S150, target difference matrix element in physical map and intermediary matrix figure is determined, and according to target difference matrix element Element determines malfunctioning node equipment and the corresponding node role of malfunctioning node equipment.
It is removed except the row and column that node device occupies in physical map and intermediary matrix figure, remaining matrix element indicates Connection relationship between two node devices, existing difference matrix element indicate node device corresponding with the matrix element Between connection relationship it is inconsistent in physical map and intermediary matrix figure, multiple nodes corresponding with target difference matrix element are set It is standby to belong to malfunctioning node equipment.
After determining malfunctioning node equipment, logical ip and physical address (MAC Address) of the equipment etc. can be obtained Information.Also, according to the corresponding node role of malfunctioning node equipment, it can use the angle to record when data center design blueprint Color information list further determines that the corresponding other information of malfunctioning node equipment, for example, according to Role Information list interior joint angle The corresponding relationship of color and node device position determines the position of malfunctioning node equipment, without passing through technical staff in data center It carries out checking the position for determining malfunctioning node equipment on the spot, facilitates the repair at the quick Develop Data center of technical staff.
Compared to existing Fault Locating Method, this embodiment scheme is not related to devising and drawing the blueprint and determining anchor in physical map It puts and is based on anchor point and gradually extend subgraph, scheme realizes that logic is more clear simply, and this programme belongs to NP (NP- completely Complete) problem, the i.e. uncertain problems of multinomial complexity, index of the computation complexity well below the prior art Grade complexity, therefore, the fault location of this programme is more efficient;And this programme gets rid of anchor point selection pair in the prior art The constraint of fault location accuracy, thus the accuracy of fault location can also be higher.
Optionally, this method further include: according to intermediary matrix figure to the connection relationship of malfunctioning node equipment in physical map into Row amendment.For example, can use the revision program write in advance, using determining malfunctioning node equipment as parameter, according to intermediate Correct node device connection relationship in matrix diagram is modified physical map, this not only grasps technical staff accurately Correct connection relationship in the data center built up between each node device, and help speed up data center and its physical map Repair process.For the less situation of connection error failure existing for data center, failure section that the present embodiment is quickly determined Point device quantity can achieve at least, so that the connection relationship reparation operation being related to is also minimum when repairing to physical map, mention High remediation efficiency
Optionally, this method further include:
According to the quantity of determining target difference matrix element, the quantity of malfunctioning node equipment is determined, use is utilized with verifying Whether meet optimal solution requirement in the quantity for the malfunctioning node equipment that the method for determining data center's failure determines, wherein optimal Solution requirement refers in the enforceable situation of this embodiment scheme, determines malfunctioning node number of devices using this embodiment scheme At least, so that reparation operation amount involved in data center and physical map is also minimum, repair process is more simple and effective.
For example, in the data center based on fat several topology network architecture buildings, it is assumed that in telephone net node equipment Port number is k, and a line indicates the connection relationship between the corresponding node device of two node roles in devising and drawing the blueprint, when event When hindering number of edges less than k/2, the quantity of the malfunctioning node equipment determined using this embodiment scheme just meets optimal solution requirement.
The technical solution of the present embodiment is layered and is grouped by node device each in the physical map to data center, It is closed according to the connection of node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint of data center System carries out parameter tags to the node device in each node group, and according to the company between node role each in devising and drawing the blueprint Relationship is connect, intermediary matrix figure, the ratio based on physical map and intermediary matrix figure are created based on each node device after parameter tags More determining malfunctioning node equipment and its node role solve malfunctioning node in existing method and determine than relatively time-consuming and accuracy Lower problem realizes the malfunctioning node for efficiently determining data center, and improves the accuracy that malfunctioning node determines;This Outside, the present embodiment technical solution situation less for connection error failure existing for data center, the failure quickly determined Node device quantity can achieve at least, so that the connection relationship reparation operation being related to is also minimum when repairing to physical map, Improve remediation efficiency.
Embodiment two
Fig. 2 is provided by Embodiment 2 of the present invention for determining the flow chart of the method for data center's failure, the present embodiment It is further progress optimization and extension on the basis of the above embodiments.As shown in Fig. 2, this method may include:
S210, node device each in the physical map of data center is divided at least one node layer according to device type, Wherein, which includes server node layer, edge switch node layer, aggregation switch node layer and core Telephone net node layer.
S220, in each node layer, according to the connection of any two node device and other node layer interior joint equipment Node device in each node layer is divided into the node group of preset quantity by the similarity between relationship.
S230, each node device of server node layer is successively subjected to parameter according to affiliated current layer interior joint group Label.
Wherein, flag parameters can be used arbitrary character and be indicated, and representation includes Chinese character, English words The combination of symbol, number and three.The node device of each node layer is marked according to affiliated current layer interior joint group When, the node device for belonging to same node group uses identical group of labelled notation.For example, node each in server node layer is set Standby successively to carry out parameter tags using number, the node device in same node group uses identical digital group number to join as label Number.
S240, according to the connection relationship between edge switch node layer and server node layer interior joint equipment, by side The node device of edge exchanger layer is carried out using group number of the server category node device being attached thereto in server node layer It marks for the first time, and the current layer interior joint group according to belonging to edge switch layer interior joint equipment successively carries out the second deutero-albumose Note, obtains the flag parameters of each node device of edge switch node layer.
Node device i.e. in edge switch node layer uses two-dimensional flag parameters, also, the parameter marked twice In the two-dimensional marker parameter and out-of-order limits, but needs the mark with the two-dimensional marker parameter for the interior joint role that devises and draw the blueprint Note rule is consistent.Illustratively, the flag parameters of the node device A in edge switch node layer are (1,3), the two-dimensional marker Flag parameters 1 in parameter can be the group number for the server category node device connecting with node device A, and flag parameters 3 can be with It is node device A in the corresponding group number of edge switch node layer;In addition, flag parameters 3 in the two-dimensional marker parameter can be with It is the group number for the server category node device being connect with node device A, flag parameters 1 can be node device A and exchange at edge The corresponding group number of machine node layer.
S250, each node device of core switch node layer is successively carried out according to affiliated current layer interior joint group Parameter tags.
For example, successively carrying out parameter to the node device of core switch node layer using number according to node group quantity Label, also, the node device in same node group uses identical digital group number as flag parameters.
S260, it is saved respectively with edge switch node layer and core switch node layer according to aggregation switch node layer The node device of aggregation switch node layer is used the edge switch class section being attached thereto by the connection relationship between point device Group number of the point device in edge switch node layer carries out first time label, and by the node device of aggregation switch node layer Second is carried out using group number of the core switch class node device being attached thereto in core switch node layer to mark, and is obtained To the flag parameters of each node device of aggregation switch node layer.
That is the flag parameters of the node device of aggregation switch node layer place one's entire reliance upon node device connected to it Flag parameters, this not only realizes the label effect to node device, but also aggregation switch section is remained in flag parameters The connection relationship of node device between point layer and adjacent node layer.For example, for the node device in aggregation switch node layer B, flag parameters are (2,3), and flag parameters 2 therein can be the edge switch class node connecting with node device B and set Standby group number, flag parameters 3 can be the group number for the core switch class node device connecting with node device B.
After carrying out parameter tags to each node device in physical map according to aforesaid operations, each node device can root Node role identical with its flag parameters is found in devising and drawing the blueprint according to the flag parameters of itself.
S270, according to the connection relationship between node role each in devising and drawing the blueprint, based on each section after parameter tags Point device creates intermediary matrix figure, wherein position of each node device in intermediary matrix figure and the position in physical map It is identical.
S280, target difference matrix element in physical map and intermediary matrix figure is determined, and according to target difference matrix element Element determines malfunctioning node equipment and the corresponding node role of malfunctioning node equipment.
The technical solution of the present embodiment is layered and is grouped by node device each in the physical map to data center, It is closed according to the connection of node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint of data center System carries out parameter tags to the node device in each node group, and according to the company between node role each in devising and drawing the blueprint Relationship is connect, intermediary matrix figure, the ratio based on physical map and intermediary matrix figure are created based on each node device after parameter tags More determining malfunctioning node equipment and its node role solve malfunctioning node in existing method and determine than relatively time-consuming and accuracy Lower problem realizes the malfunctioning node for efficiently determining data center, and improves the accuracy that malfunctioning node determines.
Embodiment three
Fig. 3 is the flow chart for the method for determining data center's failure that the embodiment of the present invention three provides, the present embodiment It is further progress optimization and extension on the basis of the above embodiments.As shown in figure 3, this method may include:
S310, node device each in the physical map of data center is divided at least one node layer according to device type.
S320, in each node layer, according to the connection of any two node device and other node layer interior joint equipment Node device in each node layer is divided into the node group of preset quantity by the similarity between relationship.
Specifically, in each node layer, according to the company of any two node device and other node layer interior joint equipment The similarity between relationship is connect, the node device in each node layer is divided into the node group of preset quantity, comprising:
In each node layer, each optional two node devices are as target node device, and according to each target section The connection relationship of point device and other node layer interior joint equipment determines the corresponding matrix subgraph of each target node device;
Matrix element in the corresponding matrix subgraph of two target node devices in same position is subjected to exclusive or calculating, really The quantity of difference matrix element in fixed two matrix subgraphs;
The company of two target node devices Yu other node layer interior joint equipment is determined according to the quantity of difference matrix element The similarity between relationship is connect, and determines whether two target node devices belong to same node group according to similarity.
Matrix subgraph in the present embodiment refers to that only including target node device itself and the connection of other node devices closes A part of physical map of system.Exclusive or between two matrix subgraphs calculate compare matrix element in same position whether phase Together.Illustratively, if in each exclusive or calculating process, the quantity of determining difference matrix element is respectively less than current setting Amount threshold, then it is assumed that the connection relationship of two objective matrix elements has similitude, can be divided into same node group.
S330, according to data center the interior joint role that devises and draw the blueprint marking convention and each node layer segmentum intercalaris The connection relationship of point device carries out parameter tags to the node device in each node group, the node device after making parameter tags It is corresponded with the node role in devising and drawing the blueprint.
S340, according to the connection relationship between node role each in devising and drawing the blueprint, based on each section after parameter tags The corresponding node role of point device creates intermediate blueprint.
The form of expression of intermediate blueprint is similar with the form of expression of data center to devise and draw the blueprint, and difference is intermediate blueprint It is to be back-calculated to obtain based on the connection relationship between each node device and each node device.In the present embodiment, data Devising and drawing the blueprint for center realizes that the network topological diagram includes but unlimited based on the network topological diagram with hierarchical structure In: fat tree topology structure chart, VL2 data center architecture figure, Google's data center network structure chart or multistage switching network framework Figure.The connection that the network topology based on hierarchical structure obtained, which devise and draw the blueprint, can clearly embody between each node role is closed System, and then clearly show the connection relationship between corresponding node device.Therefore, the intermediate blueprint created also has layer The network structure form of secondaryization can clearly embody the connection relationship of physical map interior joint equipment.
Fig. 4 a shows a kind of schematic diagram of intermediate blueprint provided in this embodiment, specifically, the node of different node layers Equipment has the stratification connection relationship of fat tree topology block diagram fonn, number mark in the flag parameters such as figure of each node device Shown in note.
S350, each node device after parameter tags is numbered to expression, and based on the number and intermediate blueprint, really Fixed intermediary matrix figure corresponding with intermediate blueprint.
Each node device is numbered, is to indicate each node device in order to convenient in matrix diagram.Number can be with Using succinct number or alpha code form.Fig. 4 b shows the schematic diagram of another intermediate blueprint provided in this embodiment, Each node device is carried out to the blueprint after digital number expression.Property indicates Fig. 4 b as an example, and opens up in Fig. 4 a The connection relationship shown does not correspond to.After node device is numbered, using the transformational relation of non-directed graph and adjacency matrix, Obtain intermediary matrix figure corresponding with intermediate blueprint.Fig. 4 c shows a kind of signal of intermediary matrix figure provided in this embodiment Figure, corresponding with the intermediate blueprint that Fig. 4 b is shown, i.e. the connection relationship between node device that Fig. 4 c and Fig. 4 b are shown is substantially identical.
S360, to each node device in physical map, using identical in intermediary matrix figure with each node device Number is indicated, and is based on the number and physical map, determines the first matrix diagram corresponding with physical map.
S370, exclusive or calculating is carried out to the matrix element in same position in the first matrix diagram and intermediary matrix figure, determined Target difference matrix element.
First matrix diagram is a kind of matrix diagram for simplifying node device after indicating relative to physical map, i.e., is replaced using number For node device MAC Address or IP address used in physical map.For same node device, in the first matrix diagram and object The number managed in figure is identical, therefore, is calculated by the exclusive or of the matrix element in same position in two matrix diagram, determines difference Matrix element, and then determine malfunctioning node equipment and the corresponding node role of malfunctioning node equipment.
S380, the corresponding malfunctioning node equipment of target difference matrix element is determined based on physical map, and be based on intermediate blueprint Determine the corresponding malfunctioning node role of target difference matrix element.
The technical solution of the present embodiment by first to each node device in the physical map of data center carry out layering with Grouping, according to node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint of data center Connection relationship carries out parameter tags to the node device in each node group;Then according to node role each in devising and drawing the blueprint Between connection relationship, intermediate blueprint is created based on each node device after parameter tags, and determining with intermediate blueprint respectively Corresponding intermediary matrix figure, and the first matrix diagram corresponding with physical map;Finally based on the first matrix diagram and intermediary matrix figure Exclusive or calculates, and determines malfunctioning node equipment and its node role, solve in existing method malfunctioning node determine than it is relatively time-consuming with And the problem that accuracy is lower, the malfunctioning node for efficiently determining data center is realized, and improve the standard that malfunctioning node determines True property.
Example IV
Fig. 5 is the structural schematic diagram for the device for determining data center's failure that the embodiment of the present invention four provides, this reality It applies example and is applicable to the case where determining the malfunctioning node of connection error in data center according to the physical map of data center.The device It can be realized, and can be integrated on an electronic device by the way of software and/or hardware.
As shown in figure 5, the device provided in this embodiment for determining data center's failure may include that node layer divides Module 410, node group division module 420, parameter tags module 430, intermediary matrix figure creation module 440 and fault determination module 450, in which:
Node layer division module 410 is divided for node device each in the physical map by data center according to device type For at least one node layer;
Node group division module 420 is used in each node layer, according to any two node device and other node layers Node device in each node layer is divided into the node of preset quantity by the similarity between the connection relationship of interior joint equipment Group;
Parameter tags module 430, for the marking convention of the interior joint role that devises and draw the blueprint according to data center, and it is each The connection relationship of node device between a node layer carries out parameter tags to the node device in each node group, makes parameter mark Node device after note with devise and draw the blueprint in node role correspond;
Intermediary matrix figure creation module 440, for according to the connection relationship between node role each in devising and drawing the blueprint, base Each node device after parameter tags creates intermediary matrix figure, wherein position of each node device in intermediary matrix figure It sets identical as the position in physical map;
Fault determination module 450, for determining the target difference matrix element in physical map and intermediary matrix figure, and according to Target difference matrix element determines malfunctioning node equipment and the corresponding node role of malfunctioning node equipment.
Optionally, at least one node layer includes server node layer, edge switch node layer, aggregation switch node Layer and core switch node layer.
Optionally, parameter tags module 430 includes:
Server node layer parameter marking unit, for working as each node device of server node layer according to affiliated Front layer interior joint group successively carries out parameter tags;
Edge switch node layer parameter tags unit, for according in edge switch node layer and server node layer The node device of edge switch layer is used the server category node device being attached thereto by the connection relationship between node device Group number in server node layer carries out first time label, and the current layer according to belonging to edge switch layer interior joint equipment Interior joint group successively carries out second and marks, and obtains the flag parameters of each node device of edge switch node layer;
Core switch node layer parameter tags unit, for by each node device of core switch node layer according to Affiliated current layer interior joint group successively carries out parameter tags;
Aggregation switch node layer parameter tags unit, for according to aggregation switch node layer respectively with edge switch Connection relationship between node layer and core switch node layer interior joint equipment, by the node device of aggregation switch node layer First time label is carried out using group number of the edge switch class node device being attached thereto in edge switch node layer, and By the node device of aggregation switch node layer using the core switch class node device being attached thereto in core switch section Group number in point layer carries out second and marks, and obtains the flag parameters of each node device of aggregation switch node layer;
Wherein, the node device of each node layer belongs to same when being marked according to affiliated current layer interior joint group The node device of one node group uses identical group of labelled notation.
Optionally, node group division module 420 includes:
Matrix subgraph determination unit, in each node layer, each optional two node devices to be as destination node Equipment, and according to the connection relationship of each target node device and other node layer interior joint equipment, determine each destination node The corresponding matrix subgraph of equipment;
Exclusive or computing unit, for by the matrix element in the corresponding matrix subgraph of two target node devices in same position Element carries out exclusive or calculating, determines the quantity of difference matrix element in two matrix subgraphs;
Node group determination unit, for determining two target node devices and other sections according to the quantity of difference matrix element Similarity between the connection relationship of point layer interior joint equipment, and determine whether two target node devices belong to according to similarity Same node group.
Optionally, intermediary matrix figure creation module 440, comprising:
Intermediate blueprint creating unit, for being based on ginseng according to the connection relationship between node role each in devising and drawing the blueprint The corresponding node role of each node device after number label creates intermediate blueprint;
Intermediary matrix figure determination unit for expression to be numbered in each node device after parameter tags, and is based on The number and intermediate blueprint, determining intermediary matrix figure corresponding with intermediate blueprint;
Correspondingly, fault determination module 450 includes:
First matrix diagram determination unit, for existing using with each node device to each node device in physical map It is identically numbered and is indicated in intermediary matrix figure, and be based on the number and physical map, determine the first square corresponding with physical map The system of battle formations;
Target difference matrix element determination unit, for the square in same position in the first matrix diagram and intermediary matrix figure Array element element carries out exclusive or calculating, determines target difference matrix element;
Failure determination unit, for determining the corresponding malfunctioning node equipment of target difference matrix element based on physical map, and The corresponding malfunctioning node role of target difference matrix element is determined based on intermediate blueprint.
Optionally, node layer division module 410, comprising:
Current SWITCH figure determination unit, the connection degree of each node device in the physical map for determining data center, Degree will be connected be less than the node device of current degree threshold value and be determined as same layer node device, and by same layer node device and its Connection relationship is removed from physical map, obtains current SWITCH figure;
Unit is repeated, for repeating the operation and removal that determine same layer node device in current SWITCH figure The operation of same layer node device and its connection relationship, until each node device is divided into corresponding node in physical map Layer.
Optionally, the device further include:
Failure proof module determines malfunctioning node equipment for the quantity according to determining target difference matrix element Quantity, it is optimal whether the quantity for the malfunctioning node equipment for utilizing the method for determining data center's failure to determine with verifying meets Solution requires.
Optionally, the device further include:
Correction module, for being modified according to intermediary matrix figure to the connection relationship of malfunctioning node equipment in physical map.
Optionally, devising and drawing the blueprint for data center is realized based on the network topological diagram with hierarchical structure.
Optionally, network topological diagram includes fat tree topology structure chart, VL2 data center architecture figure, data center, Google net Network structure chart or multistage switching network architecture diagram.
For determining that any embodiment of that present invention can be performed in the device of data center's failure provided by the embodiment of the present invention The provided method for determining data center's failure, has the corresponding functional module of execution method and beneficial effect.This reality The content for applying not detailed description in example can be with reference to the content description in embodiment of the present invention method.
Embodiment five
Fig. 6 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention five provides.Fig. 6, which is shown, to be suitable for being used in fact The block diagram of the example electronic device 512 of existing embodiment of the present invention.The electronic equipment 512 that Fig. 6 is shown is only an example, Should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in fig. 6, electronic equipment 512 is showed in the form of universal electronic device.The component of electronic equipment 512 can wrap Include but be not limited to: one or more processor 516, storage device 528 connect different system components (including storage device 528 With processor 516) bus 518.
Bus 518 indicates one of a few class bus structures or a variety of, including storage device bus or storage device control Device processed, peripheral bus, graphics acceleration port, processor or total using the local of any bus structures in a variety of bus structures Line.For example, these architectures include but is not limited to industry standard architecture (Industry Subversive Alliance, ISA) bus, microchannel architecture (Micro Channel Architecture, MAC) bus is enhanced Isa bus, Video Electronics Standards Association (Video Electronics Standards Association, VESA) local are total Line and peripheral component interconnection (Peripheral Component Interconnect, PCI) bus.
Electronic equipment 512 typically comprises a variety of computer system readable media.These media can be it is any can be by The usable medium that electronic equipment 512 accesses, including volatile and non-volatile media, moveable and immovable medium.
Storage device 528 may include the computer system readable media of form of volatile memory, such as arbitrary access Memory (Random Access Memory, RAM) 530 and/or cache memory 532.Electronic equipment 512 can be into one Step includes other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example, it stores System 534 can be used for reading and writing immovable, non-volatile magnetic media (Fig. 6 do not show, commonly referred to as " hard disk drive "). Although being not shown in Fig. 6, the disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided, And to removable anonvolatile optical disk, such as CD-ROM (Compact Disc Read-Only Memory, CD-ROM), Digital video disk (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical mediums) read-write light Disk drive.In these cases, each driver can pass through one or more data media interfaces and 518 phase of bus Even.Storage device 528 may include at least one program product, which has one group of (for example, at least one) program mould Block, these program modules are configured to perform the function of various embodiments of the present invention.
Program/utility 540 with one group of (at least one) program module 542 can store in such as storage dress It sets in 528, such program module 542 includes but is not limited to operating system, one or more application program, other program moulds It may include the realization of network environment in block and program data, each of these examples or certain combination.Program module 542 usually execute function and/or method in embodiment described in the invention.
Electronic equipment 512 (such as keyboard, can also be directed toward terminal, display 524 with one or more external equipments 514 Deng) communication, can also be enabled a user to one or more terminal interact with the electronic equipment 512 communicate, and/or with make Any terminal that the electronic equipment 512 can be communicated with one or more of the other computing terminal (such as network interface card, modem Etc.) communication.This communication can be carried out by input/output (I/O) interface 522.Also, electronic equipment 512 can also lead to Cross network adapter 520 and one or more network (such as local area network (Local Area Network, LAN), wide area network (Wide Area Network, WAN) and/or public network, such as internet) communication.As shown in fig. 6, network adapter 520 It is communicated by bus 518 with other modules of electronic equipment 512.It should be understood that although not shown in the drawings, can be set in conjunction with electronics Standby 512 use other hardware and/or software module, including but not limited to: microcode, terminal driver, redundant processor, outside Disk drive array, disk array (Redundant Arrays of Independent Disks, RAID) system, tape drive Dynamic device and data backup storage system etc..
The program that processor 516 is stored in storage device 528 by operation, thereby executing various function application and number According to processing, such as realize the method provided by any embodiment of the invention for determining data center's failure, this method can be with Include:
Node device each in the physical map of data center is divided at least one node layer according to device type;
In each node layer, according to the connection relationship of any two node device and other node layer interior joint equipment it Between similarity, the node device in each node layer is divided into the node group of preset quantity;
According to the marking convention of the interior joint role that devises and draw the blueprint of the data center and the intermediate node of each node layer The connection relationship of equipment carries out parameter tags to the node device in each node group, node device after making parameter tags with It is described devise and draw the blueprint in node role correspond;
According to the connection relationship between node role each in described devise and draw the blueprint, based on each node after parameter tags Equipment create intermediary matrix figure, wherein position of each node device in the intermediary matrix figure in the physical map Position it is identical;
Determine the target difference matrix element in the physical map and the intermediary matrix figure, and according to the target difference Matrix element determines malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment.
Embodiment six
The embodiment of the present invention six additionally provides a kind of computer readable storage medium, is stored thereon with computer program, should Such as method provided by any embodiment of the invention for determining data center's failure is realized when program is executed by processor, it should Method may include:
Node device each in the physical map of data center is divided at least one node layer according to device type;
In each node layer, according to the connection relationship of any two node device and other node layer interior joint equipment it Between similarity, the node device in each node layer is divided into the node group of preset quantity;
According to the marking convention of the interior joint role that devises and draw the blueprint of the data center and the intermediate node of each node layer The connection relationship of equipment carries out parameter tags to the node device in each node group, node device after making parameter tags with It is described devise and draw the blueprint in node role correspond;
According to the connection relationship between node role each in described devise and draw the blueprint, based on each node after parameter tags Equipment create intermediary matrix figure, wherein position of each node device in the intermediary matrix figure in the physical map Position it is identical;
Determine the target difference matrix element in the physical map and the intermediary matrix figure, and according to the target difference Matrix element determines malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD- ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device Using or it is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal, Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++, It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion Divide and partially executes or executed on remote computer or terminal completely on the remote computer on the user computer.It is relating to And in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or extensively Domain net (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as provided using Internet service Quotient is connected by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation, It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.

Claims (13)

1. a kind of method for determining data center's failure characterized by comprising
Node device each in the physical map of data center is divided at least one node layer according to device type;
In each node layer, according between any two node device and the connection relationship of other node layer interior joint equipment Node device in each node layer is divided into the node group of preset quantity by similarity;
According to node device between the marking convention and each node layer of the interior joint role that devises and draw the blueprint of the data center Connection relationship, parameter tags are carried out to the node device in each node group, node device after making parameter tags with it is described Node role in devising and drawing the blueprint corresponds;
According to the connection relationship between node role each in described devise and draw the blueprint, based on each node device after parameter tags Create intermediary matrix figure, wherein position of each node device in the intermediary matrix figure and the position in the physical map It sets identical;
Determine the target difference matrix element in the physical map and the intermediary matrix figure, and according to the target difference matrix Element determines malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment.
2. the method according to claim 1, wherein at least one described node layer include server node layer, Edge switch node layer, aggregation switch node layer and core switch node layer.
3. according to the method described in claim 2, it is characterized in that, according to the data center the interior joint role that devises and draw the blueprint Marking convention and each node layer between node device connection relationship, in each node group node device carry out Parameter tags, comprising:
Each node device of the server node layer is successively subjected to parameter tags according to affiliated current layer interior joint group;
It, will be described according to the connection relationship between the edge switch node layer and the server node layer interior joint equipment Group of the node device of edge switch layer using the server category node device being attached thereto in the server node layer Number first time label is carried out, and the current layer interior joint group according to belonging to the edge switch layer interior joint equipment successively carries out Second of label, obtains the flag parameters of each node device of the edge switch node layer;
Each node device of the core switch node layer is successively subjected to parameter according to affiliated current layer interior joint group Label;
According to the aggregation switch node layer respectively with the edge switch node layer and the core switch node layer Connection relationship between interior joint equipment hands over the node device of the aggregation switch node layer using the edge being attached thereto Group number of class of the changing planes node device in the edge switch node layer carries out first time label, and by the aggregation switch The node device of node layer is using the core switch class node device being attached thereto in the core switch node layer Group number carries out second and marks, and obtains the flag parameters of each node device of the aggregation switch node layer;
Wherein, the node device of each node layer belongs to same section when being marked according to affiliated current layer interior joint group The node device of point group uses identical group of labelled notation.
4. the method according to claim 1, wherein described in each node layer, according to any two node Similarity between equipment and the connection relationship of other node layer interior joint equipment divides the node device in each node layer For the node group of preset quantity, comprising:
In each node layer, each optional two node devices are set as target node device, and according to each destination node The standby connection relationship with other node layer interior joint equipment determines the corresponding matrix subgraph of each target node device;
Matrix element in the corresponding matrix subgraph of two target node devices in same position is subjected to exclusive or calculating, determines two The quantity of difference matrix element in a matrix subgraph;
Described two target node devices and other node layer interior joint equipment are determined according to the quantity of the difference matrix element Connection relationship between similarity, and determine whether described two target node devices belong to same section according to the similarity Point group.
5. the method according to claim 1, wherein according between node role each in described devise and draw the blueprint Connection relationship creates intermediary matrix figure based on each node device after parameter tags, comprising:
Each node according to the connection relationship between node role each in described devise and draw the blueprint, after marking based on the parameter The corresponding node role of equipment creates intermediate blueprint;
Expression is numbered in each node device after the parameter tags, and is based on the number and the intermediate blueprint, really Fixed intermediary matrix figure corresponding with the intermediate blueprint;
Correspondingly, determining the target difference matrix element in the physical map and the intermediary matrix figure, and according to the target Difference matrix element determines malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment, comprising:
To each node device in the physical map, using with each node device in the intermediary matrix figure identical volume It number is indicated, and is based on the number and the physical map, determine the first matrix diagram corresponding with the physical map;
Exclusive or calculating is carried out to the matrix element in same position in first matrix diagram and the intermediary matrix figure, determines mesh Mark difference matrix element;
The corresponding malfunctioning node equipment of the target difference matrix element is determined based on the physical map, and based on described intermediate blue Figure determines the corresponding malfunctioning node role of the target difference matrix element.
6. the method according to claim 1, wherein each node device is pressed in the physical map by data center At least one node layer is divided into according to device type, comprising:
Connection degree is less than current degree threshold value by the connection degree for determining each node device in the physical map of the data center Node device be determined as same layer node device, and by the same layer node device and its connection relationship from the physical map Removal, obtains current SWITCH figure;
It repeats the operation for determining the same layer node device in the current SWITCH figure and removes the same node layer The operation of equipment and its connection relationship, until each node device is divided into corresponding node layer in the physical map.
7. the method according to claim 1, wherein the method also includes:
According to the quantity of the determining target difference matrix element, the quantity of the malfunctioning node equipment is determined, to verify benefit Whether the quantity of the malfunctioning node equipment determined with the method for determining data center's failure, which meets optimal solution, is wanted It asks.
8. the method according to claim 1, wherein the method also includes:
It is modified according to connection relationship of the intermediary matrix figure to malfunctioning node equipment in the physical map.
9. any method in -8 according to claim 1, which is characterized in that described to devise and draw the blueprint based on having stratification knot The network topological diagram of structure feature is realized.
10. according to the method described in claim 9, it is characterized in that, the network topological diagram include fat tree topology structure chart, VL2 data center architecture figure, Google's data center network structure chart or multistage switching network architecture diagram.
11. a kind of for determining the device of data center's failure characterized by comprising
Node layer division module is divided at least one according to device type for node device each in the physical map by data center A node layer;
Node group division module is used in each node layer, according to any two node device and other node layer interior joints Node device in each node layer is divided into the node group of preset quantity by the similarity between the connection relationship of equipment;
Parameter tags module, for the marking convention of the interior joint role that devises and draw the blueprint according to the data center and each The connection relationship of node device between node layer carries out parameter tags to the node device in each node group, makes parameter tags Node device afterwards is corresponded with the node role in described devise and draw the blueprint;
Intermediary matrix figure creation module, for being based on according to the connection relationship between node role each in described devise and draw the blueprint Each node device after parameter tags creates intermediary matrix figure, wherein each node device is in the intermediary matrix figure Position is identical as the position in the physical map;
Fault determination module, for determining the target difference matrix element in the physical map and the intermediary matrix figure, and root Malfunctioning node equipment and the corresponding node role of the malfunctioning node equipment are determined according to the target difference matrix element.
12. a kind of electronic equipment characterized by comprising
One or more processors;
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real The now method for determining data center's failure as described in any in claim 1-10.
13. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor The method for determining data center's failure as described in any in claim 1-10 is realized when execution.
CN201910110034.9A 2019-02-11 2019-02-11 Method, apparatus, electronic device, and medium for determining data center failure Active CN109688758B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910110034.9A CN109688758B (en) 2019-02-11 2019-02-11 Method, apparatus, electronic device, and medium for determining data center failure

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910110034.9A CN109688758B (en) 2019-02-11 2019-02-11 Method, apparatus, electronic device, and medium for determining data center failure

Publications (2)

Publication Number Publication Date
CN109688758A true CN109688758A (en) 2019-04-26
CN109688758B CN109688758B (en) 2022-04-12

Family

ID=66195622

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910110034.9A Active CN109688758B (en) 2019-02-11 2019-02-11 Method, apparatus, electronic device, and medium for determining data center failure

Country Status (1)

Country Link
CN (1) CN109688758B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110519109A (en) * 2019-09-26 2019-11-29 中国工商银行股份有限公司 The associated method, apparatus of probe node calculates equipment and medium
CN112511341A (en) * 2020-11-14 2021-03-16 鹏城实验室 Network automation fault positioning method, terminal and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546852A (en) * 2011-12-13 2012-07-04 西安交通大学 Address automatic configuration method of fault-tolerant data center network
CN103269387A (en) * 2013-05-21 2013-08-28 中国科学院计算技术研究所 Automatic configuration method for data center network addresses
US20140344438A1 (en) * 2010-12-14 2014-11-20 Microsoft Corporation Generic and automatic address configuration for data center networks
CN106992604A (en) * 2017-05-31 2017-07-28 泉州创先力智能科技有限公司 A kind of fault location and visual presentation device based on automation equipment virtual connections
CN107046481A (en) * 2017-04-18 2017-08-15 国网福建省电力有限公司 A kind of information system integrated network management system comprehensive analysis platform
CN108959666A (en) * 2017-05-17 2018-12-07 中国科学院微电子研究所 Method of designing integrated circuit and device, chip layout decomposition and color method and device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140344438A1 (en) * 2010-12-14 2014-11-20 Microsoft Corporation Generic and automatic address configuration for data center networks
CN102546852A (en) * 2011-12-13 2012-07-04 西安交通大学 Address automatic configuration method of fault-tolerant data center network
CN103269387A (en) * 2013-05-21 2013-08-28 中国科学院计算技术研究所 Automatic configuration method for data center network addresses
CN107046481A (en) * 2017-04-18 2017-08-15 国网福建省电力有限公司 A kind of information system integrated network management system comprehensive analysis platform
CN108959666A (en) * 2017-05-17 2018-12-07 中国科学院微电子研究所 Method of designing integrated circuit and device, chip layout decomposition and color method and device
CN106992604A (en) * 2017-05-31 2017-07-28 泉州创先力智能科技有限公司 A kind of fault location and visual presentation device based on automation equipment virtual connections

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110519109A (en) * 2019-09-26 2019-11-29 中国工商银行股份有限公司 The associated method, apparatus of probe node calculates equipment and medium
CN112511341A (en) * 2020-11-14 2021-03-16 鹏城实验室 Network automation fault positioning method, terminal and storage medium
CN112511341B (en) * 2020-11-14 2023-04-07 鹏城实验室 Network automation fault positioning method, terminal and storage medium

Also Published As

Publication number Publication date
CN109688758B (en) 2022-04-12

Similar Documents

Publication Publication Date Title
US20140344438A1 (en) Generic and automatic address configuration for data center networks
US9317327B2 (en) Computing infrastructure planning
CN105446896B (en) The buffer memory management method and device of MapReduce application
CN105446793A (en) Method and device for migrating virtual assets
EP3320653A1 (en) Parallelized network traffic flow availability simulation using stochastic process and traffic engineering algorithms
CN107506484B (en) Operation and maintenance data association auditing method, system, equipment and storage medium
CN104348643A (en) Visual topology generating method and device
CN109688758A (en) For determining method, apparatus, electronic equipment and the medium of data center's failure
CN108494607A (en) The design method and system of big double layer network framework based on container
US8996341B2 (en) Generating and evaluating expert networks
US10606975B2 (en) Coordinates-based generative adversarial networks for generating synthetic physical design layout patterns
CN109446202A (en) Identifier allocation method, device, server and storage medium
CN109818796B (en) Data center construction method and device, electronic equipment and medium
CN109861863A (en) The connecting fault of data center determines method, apparatus, electronic equipment and medium
CN109861869B (en) Configuration file generation method and device
CN109756381A (en) Data center's Fault Locating Method, device, electronic equipment and medium
US10546096B2 (en) Semiconductor package via stack checking
CN101505241A (en) Method and apparatus for generating test instances
CN104572687A (en) Recognition method and device of microblogging key users
CN110430140A (en) Path processing method, device, equipment and storage medium
CN110061917A (en) Creation method, device, the network equipment and the storage medium of interface group
CN106209420B (en) A kind of method and electronic equipment of location data forwarding service failure
CN110391934B (en) Network equipment labeling method and device
US9158871B2 (en) Graph modeling systems and methods
CN108768735B (en) Bipartite graph sampling method and device for test bed topological structure

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant