CN106775953A - Realize the method and system of OpenStack High Availabitities - Google Patents
Realize the method and system of OpenStack High Availabitities Download PDFInfo
- Publication number
- CN106775953A CN106775953A CN201611265200.5A CN201611265200A CN106775953A CN 106775953 A CN106775953 A CN 106775953A CN 201611265200 A CN201611265200 A CN 201611265200A CN 106775953 A CN106775953 A CN 106775953A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- node
- control node
- request
- control
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45562—Creating, deleting, cloning virtual machine instances
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses the method and system for realizing OpenStack High Availabitities, the control system that it is based on includes scheduler, two configurations identical control node and Duo Tai calculate nodes, scheduler receives the virtual machine treatment request of client, according to two control nodes each calculate node loading condition, determine that the small control node of load capacity is target control node, virtual machine treatment request is sent to target control node, target control node is according to many resource service conditions of calculate node, determine target computing nodes, and by virtual machine treatment request be sent to target computing nodes come response data request.When any one control node or calculate node break down, the control node of other normal operations or the task of calculate node taking over fault node, and at least there are two independent databases in each control node, the data that each database is preserved are identical, if a database failure, other databases take over this database tasks.
Description
Technical field
The present invention relates to OpenStack cloud computing platforms field, during more particularly to one kind realizes OpenStack control systems
The method and apparatus of the database High Availabitity in any one control node, a kind of method for realizing OpenStack High Availabitities with
System.
Background technology
At present, with Internet of Things, the rise of Mobile solution, information system needs data volume to be processed to compare geometry in the past
The lifting of level, therefore, the requirement of the application of enterprise to database be no longer it is simple it is sometime interior can process " increase, delete,
Change, look into " etc. request efficiency and the performance such as response time;Also need to solve, when database runs into failure, to user's
Service can not be interrupted, and database should (database be successfully processed number higher within a certain period of time with handling capacity higher
According to the data of amount).
In face of the problem that above-mentioned needs are solved, increasing enterprise processes the business of enterprise using OpenStack, by
Many databases in its control node can be constituted a data-base cluster, when any database in OpenStack
During failure, other databases in data-base cluster can take over the task of this database to provide the user service, because
This can provide the user service with continual;Additionally, when client application is loaded and increased, new database is added
To data-base cluster, realize to the extending transversely of data-base cluster so that data-base cluster has handling capacity higher to data,
And then meet the application of enterprise.
In the prior art, data-base cluster is realized using Share-Disk frameworks, database is total in this data-base cluster
Enjoy a database and carry out data storage.Additionally, there is two ways, Yi Zhongfang in the data-base cluster that Share-Disk frameworks are realized
Formula is:Only have a database externally to provide service in cluster, (this kind of mode is referred to as list to other databases used as redundant server
It is living), when the database for externally providing service breaks down, redundant server externally provides service;Another way is:Data
All of database all externally provides service (this kind of mode into claim dual-active) in the cluster of storehouse, and typical product is RAC.
The content of the invention
Inventor has found in research process, two kinds of data-base cluster sides that Share-Disk frameworks are provided in the prior art
All there are some problems in formula.In former data-base cluster, due to only having a database external in cluster in the same time
Service is provided, so as to cause the serious waste of hardware resource, it is impossible to lift the performance that data-base cluster externally provides service;Rear
It is technical very high due to RAC, it is therefore desirable to which that level people higher carrys out operational system in a kind of data-base cluster, also,
RAC needs associated client application to use, if RAC is mismatched with client application, causes the number of RAC
According to the dramatic decline of storehouse performance;Additionally, the data-base cluster that both modes are realized, cluster shares a database to store number
According to storage aspect has Single Point of Faliure, i.e., when shared data bank goes wrong, whole data-base cluster cannot normal work.
In view of this, the main object of the present invention is to overcome two kinds of data that Share-Disk frameworks are provided in the prior art
The shortcoming of storehouse cluster, realization meets the High Availabitity data-base cluster scheme of enterprise's application, and then realizes for enterprise OpenStack
High Availabitity.
Therefore, technical proposal that the invention solves the above-mentioned problems is:
A kind of method for realizing database High Availabitity, it is characterised in that methods described is applied to the control based on OpenStack
In any one control node of system processed, the control system includes two control nodes and multiple calculate nodes;The method
Including:
Receive the virtual machine treatment request that different clients send;
The target computing nodes of the response virtual machine treatment request are determined according to default load-balancing algorithm;
Virtual machine treatment request is sent to the target computing nodes, so that the target computing nodes perform institute
State request of data.
Preferably, a kind of method for realizing database High Availabitity, it is characterised in that also include:
The virtual machine processing information in the virtual machine treatment request is obtained, the virtual machine processing information includes:Virtually
Machine creates information, deletes information or modification information;
The client and the corresponding virtual machine processing information are stored into the calculate node table of database.
Preferably, a kind of method for realizing database High Availabitity, it is characterised in that also include:
Monitor the running status of the virtual machine in the multiple calculate node;
The resource service condition of each calculate node is counted according to the running status of the virtual machine;
The resource service condition of each calculate node is stored in the calculate node table of database.
Preferably, the target for determining the response virtual machine treatment request according to default load-balancing algorithm calculates section
Point, including:
Resource is defined as using minimum calculate node to respond the target computing nodes that the virtual machine processes request.
The present invention provides a kind of device for realizing database High Availabitity, it is characterised in that described device is integrated in and is based on
In any one control node of the control system of OpenStack, the control system includes two control nodes and multiple meters
Operator node;Described device includes:
Request unit is received, the virtual machine for receiving different clients transmission processes request;
Target computing nodes unit is determined, for determining that the response virtual machine is processed according to default load-balancing algorithm
The target computing nodes of request;
Request unit is sent, for virtual machine treatment request to be sent to the target computing nodes, so as to described
Target computing nodes perform the request of data.
Preferably, the device for realizing database High Availabitity also includes:
Virtual machine processing information unit is obtained, for obtaining the virtual machine processing information in the virtual machine treatment request,
The virtual machine processing information includes:Virtual machine creating information, deletion information or modification information;
Storage virtual machine processing information unit, for by the client and the corresponding virtual machine processing information
Store into the calculate node table of database.
Preferably, control node includes at least two databases independent mutually, the storage virtual machine processing information list
Unit includes:
Memory module, for the client to be stored to any one database with corresponding virtual machine processing information
In calculate node table;
Replication module, the calculating for the client to be copied to other databases with corresponding virtual machine processing information
In node table.
Preferably, the device for realizing database High Availabitity also includes:
Monitoring unit, the running status for monitoring the virtual machine in the multiple calculate node;
Statistic unit, the resource service condition of each calculate node is counted for the running status according to the virtual machine;
Storage unit, the calculate node table for the resource service condition of each calculate node to be stored in database
In.
Preferably, a kind of device for realizing database High Availabitity, wherein it is determined that target computing nodes unit, specifically
For resource to be defined as using minimum calculate node to respond the target computing nodes that the virtual machine processes request.
The present invention also provides a kind of method for realizing OpenStack High Availabitities, it is characterised in that methods described is applied to base
In the control system of OpenStack, the control system based on OpenStack includes:Scheduler, two configuration identical controls
Node processed and multiple calculate nodes, the method include:
The scheduler receives the virtual machine treatment request that client sends;
The loading condition of the scheduler each control node described in, it is determined that responding the virtual machine treatment request
One target control node;
Virtual machine treatment request is sent to the target control node by the scheduler;
The target control node receives the virtual machine treatment request of the scheduler dispatches, according to default load balancing
Algorithm determines the target computing nodes of response virtual machine treatment request, and virtual machine treatment request is sent to described
Target computing nodes;
The target computing nodes perform the request of data.
Preferably, the loading condition of the scheduler each control node described in, it is determined that responding at the virtual machine
The target control node of request is managed, including:
The scheduler obtains the loading condition of each control node from internal memory, each control node it is negative
Load situation is stored in the respective internal memory of the control node;
The loading condition of the scheduler each control node described in, by the control section corresponding to minimal negative carrying capacity
Point, is defined as responding the target control node of the virtual machine treatment request.
The present invention also provides a kind of system for realizing OpenStack High Availabitities, it is characterised in that the system integration is in base
In the control system of OpenStack, the control system based on OpenStack includes:Scheduler, two configuration identical controls
Node processed and multiple calculate nodes, the scheduler includes receiving unit, determines target control node unit and transmitting element, institute
State receiving unit and process request for receiving the virtual machine of client transmission;It is described determination target control node unit, for according to
According to the loading condition of each control node, it is determined that responding the target control node of the virtual machine treatment request;The hair
Unit is sent, for virtual machine treatment request to be sent into the target control node;
The device of integrated above-mentioned any one database High Availabitity in the control node;
The calculate node is used to perform the request of data.
Compared with prior art, the invention has the advantages that:
The present invention has two in data-base cluster High Availabitity in realizing OpenStack control nodes in this data-base cluster
Platform database, under normal circumstances, the database purchase institute in two database external offer services, and two databases simultaneously
Have using user's partial data, when wherein one database breaks down, another database taking over fault server is appointed
Business, to provide unremitting service using user, while overcome data-base cluster in the prior art has list in terms of storage
The shortcoming of point failure;Also, this data-base cluster is not required application program, any application program can all use this database
Cluster, overcomes in the prior art, and the suitable application program of RAC product neededs is come the shortcoming that uses.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of method flow diagram for realizing database High Availabitity in the present invention;
Fig. 2 is a kind of method flow diagram for realizing OpenStack High Availabitities in the present invention;
Fig. 3 is a kind of schematic device for realizing database High Availabitity in the present invention;
Fig. 4 is a kind of schematic device for realizing OpenStack High Availabitities in the present invention.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
The present invention provides a kind of method for realizing database High Availabitity, and the method is applied in OpenStack control systems
In any one control node, wherein, OpenStack control systems include two control nodes with multiple calculate nodes.Every
At least include two independent databases in individual control node, and the data of the database preservation in control node are identical, when
After one database breaks down, no matter this failure is operation flow, physical facility, or IT software/hardware failures, control node
In other databases continue as control node provide service, to realize the High Availabitity of database in control node.According to difference
Client is asked the different disposal of virtual machine, and control node can create virtual machine in calculate node, can also be saved to calculating
Virtual machine on point is deleted and changed, to meet client demand.
The present invention realizes the process of database High Availabitity in any one control node of OpenStack control systems
For:First, control node receives the virtual machine treatment request sent from different clients and (creates virtual machine, deletes virtual machine
Or modification virtual machine);Then, determine that the virtual machine that customer in response end sends processes request according to default load-balancing algorithm
Target computing nodes;Finally, the treatment request of virtual machine is sent to target computing nodes, so that target computing nodes are responded
The treatment request of this virtual machine.
The present invention realizes the basis of database High Availabitity in any one control node of OpenStack control systems
On, a kind of method for realizing OpenStack High Availabitities is also achieved, wherein, OpenStack High Availabitities are based on OpenStack
Control system, OpenStack control systems include scheduler, two configuration identical control nodes and multiple calculate node.
The present invention realizes that the process of OpenStack High Availabitities is:First, scheduler receives the virtual machine treatment from different clients and asks
Ask;Then, scheduler determines a target control node to ring according to the loading condition of each control node in control system
The virtual machine treatment request that should now receive, and the treatment request of this virtual machine is sent to the target control node of determination;Finally,
Target control node realizes database High Availabitity according to the present invention in any one control node of OpenStack control systems
Method, target control node determine response now virtual machine treatment request target computing nodes, so as to this target calculate
Node processes request to perform according to virtual machine.
In order that those skilled in the art more fully understand the present invention program, with reference to the accompanying drawings and detailed description to this
Invention is described in further detail.
Embodiment 1
In order that those skilled in the art more fully understand that one kind of the invention realizes OpenStack high availability methods, below
Carried out first to the present invention in any one control node in OpenStack control systems, the step of realize database High Availabitity
It is discussed in detail, it is necessary to do following preparation before database High Availabitity step is performed.
Using SaltStack instruments in two control nodes of OpenStack Platform deployments and many calculate nodes, every control
At least include two databases in node processed, the configuration of each database is identical, come using load-balanced server (HAProxy)
It is the virtual machine treatment request distribution calculate node of each client.It is specifically divided into the following steps:
(1) system initialization, various API services and inner working components are installed in control node and prepare yum sources;
(2) in control node installation database and related software bag;
(3) data-base cluster is set, and specifically includes:Each database binglog forms are configured for ROW;Configuration database
Cluster name;Each database hostname and IP address information in configuration database cluster;First use " wsrep-new-cluster "
Parameter starts a database, and remaining database is started after starting successfully;
(4) installed and configuration load equalization server (HAProxy) in control node;
(5) test connects database and checks data-base cluster state.
Fig. 1, is a kind of method flow diagram for realizing database High Availabitity provided in an embodiment of the present invention, and specific steps can be with
Including:
Step S100:Virtual machine treatment request is sent to control node by different clients.
In this step, any one client in different clients is required for calculating section according to business demand
The business on virtual machine is processed on point, for example, creating virtual machine in calculate node, deleting virtual machine or modification virtual machine
Etc. business.In OpenStack cloud platforms, calculate node has multiple, therefore, respond the virtual machine treatment that each client sends
The calculate node of request, is required for being chosen by the control node in OpenStack control systems.
Control node receives the virtual machine treatment request of different clients transmission first, and then choosing one, to respond this virtual
The calculate node of machine treatment request, the calculate node being finally selected performs the request of data of virtual machine.
Step S101:Control node receives the virtual machine treatment request that different clients send.
The sequencing of the virtual machine treatment request sent according to different clients, the load balancing service in control node
Device receives the virtual machine treatment request of all clients successively according to sequencing, and according to this sequentially for each client is selected
Take the target computing nodes for responding the treatment request of this virtual machine.
Step S102:Control node determines the mesh of the response virtual machine treatment request according to default load-balancing algorithm
Mark calculate node.
In upper step, the load-balanced server in control node have received different clients successively according to sequencing
The virtual machine treatment request at end, the load-balancing algorithm in load-balanced server, according to described sequencing, according to now
The load capacity of each calculate node in control system, is that each client distribution one responds the treatment request of this virtual machine
Target computing nodes.By taking a client as an example, control node determines a target computing nodes to respond this for this client
The specific steps of the virtual machine treatment request of client include:Step A1~step A3.
Step A1:Pacemaker in control node monitors virtual machine in the multiple calculate nodes in now control system
Running status.
In actual production environment, the calculate node in control system is likely encountered hardware fault or software fault, its
In, hardware fault, such as power down, internal memory and cpu fault etc.;Software fault is mainly the failure in terms of operating system, for example,
Kernel panic services are abnormal to service abnormal etc. with Neutron-ovs-agent.In order to determine the virtual machine at customer in response end
The target computing nodes of request are processed, it is necessary to control node monitors the running status of virtual machine in each calculate node.
Generally, in OpenStack control systems all installed in control node and each calculate node pacemaker and
The control node of corosync, OpenStack control system is monitored in each calculate node by pacemaker and corosync
The running status of virtual machine, but, the corosync of control system can only at most monitor 16 calculate nodes, and general
In OpenStack private clounds, the number of calculate node all can be more than 16.In order to overcome the shortcoming of calculate node limited amount,
The embodiment of the present invention installs pacemaker-remote to replace pacemaker and corosync in each calculate node, now,
Control node can monitor the pacemaker-remote in each calculate node come in monitoring calculate node by pacemaker
Whether virtual machine is normally running.
The mode that whether virtual machine normally runs in control node each calculate node of monitoring of control system has various, except
The pacemaker-remote in each calculate node is monitored by the pacemaker in above-mentioned control node to judge each meter
Outside whether virtual machine normally runs in operator node, the pacemaker in control node can also be by monitoring each calculate node
In the process such as nova-compute, neurton-ovs-agent and libvirt, it might even be possible to by each calculate node
Upper startup virtual machine, and then whether virtual machine normally runs in obtaining each calculate node.
In the embodiment of the present invention, by the pacemaker in control node, in real time in each calculate node
Whether pacemaker-remote sends the normal operation information of virtual machine, if in the time of regulation, in control node
Pacemaker receives the feedback information of pacemaker-remote in calculate node, and virtual machine is just in representing this calculate node
Often operation, if in the time of regulation, the pacemaker in control node does not receive pacemaker- in calculate node
The feedback information of remote, virtual machine occurs in that failure in representing this calculate node.
Within the time of regulation, Pacemaker will be isolated and be recovered to the calculate node for not receiving feedback information.
Isolation is exactly virtual machine removal OpenStack in calculate node by failure;Recovery refers to by the virtual of this calculation of fault node
Machine is moved in the calculate node of other normal operations;The concrete operations of isolation are:Pacemaker is using the fence- in it
Come virtual machine in the calculate node for closing failure, then, the fence-compute in Pacemaker is supervised ipmilan always
Control in the calculate node of this failure virtual machine whether down, when the fence-compute in Pacemaker determines this failure
In calculate node after virtual machine down, the operation for recovering is performed:Nava in control node performs host-evacuate by down
Calculate node in virtual machine (vm) migration to other normal operations calculate node, it is virtual in above-mentioned migration calculation of fault node
The time that machine needs may be more long, and in order to migrate the virtual machine of calculation of fault node faster, the embodiment of the present invention is called
Virtual machine indicia in the calculate node of failure is directly down by the force-down API of Liberty versions, so as to faster
Migrate the virtual machine on this calculation of fault node.
Now, the middle virtual machine of calculate node failure has been isolated, and the virtual machine of failure has been transferred in this calculate node
Onto the calculate node of other normal operations, then, pacemaker to the calculate node of each normal operation, sends one respectively
Individual Current resource service condition information, the resource service condition information refers to and is currently running virtual machine in each calculate node
Quantity, be sent to for the information of resource service condition by and the content that every virtual machine is run, each calculate node
Pacemaker, pacemaker will count the resource service condition information that each calculate node is replied, and in any one database
Calculate node table (compute_nodes tables) in, the corresponding load capacity of each calculate node is preserved, so that load balancing takes
The information that business device is preserved according to calculate node table (compute_nodes tables), customer in response is determined from these calculate nodes
One target computing nodes of end virtual machine treatment request.
Step A2:Load-balanced server counts each respectively according to the running status of virtual machine in each calculate node
The resource service condition of calculate node.
Load-balanced server calls calculate node table (compute_ from any one database of control node
Nodes tables), the Current resource service condition of each calculate node in this control system, i.e., each calculate node are obtained from this table
In the quantity of virtual machine that is currently running, so as to the quantity according to the virtual machine run in each calculate node number determine
One target computing nodes of customer in response end virtual machine treatment request.
Step A3:According to the resource service condition that each calculate node for obtaining is counted in step A2, resource is used minimum
Calculate node be defined as respond this client virtual machine treatment request calculate node.
According to the resource service condition that each calculate node for obtaining is counted in step A2, i.e., run in current each calculate node
Virtual machine quantity, resource is found from this information using minimum corresponding calculate node, this resource is right using minimum institute
The calculate node answered is defined as responding the target computing nodes of this client virtual machine treatment request.
Step A1~step A3, is discussed in detail by taking the virtual machine treatment request that a client sends as an example, by control
Whether virtual machine normally runs during pacemaker in node monitors each calculate node in now control system, then, load
Equalization server is that this client distributes a process for calculate node.For other clients that load-balanced server is received
Virtual treatment request, also according to the process of step A1~step A3, be followed successively by each client determine one respond this visitor
Family end virtually processes the target computing nodes of request.
When load-balanced server is that a client determines a target for responding the treatment request of this client virtual machine
After calculate node, load-balanced server is by the virtual machine treatment request of this client, this client and responds at this virtual machine
Manage the target computing nodes of request, the calculate node table of any one database of the corresponding preservation in this control node
In (compute_nodes tables).For example, for No. 1 client, sending one to load-balanced server needs in calculate node
One request of virtual machine of upper establishment, and load-balanced server is to that define No. 2 calculate nodes to respond this client
Request, now, the information of " No. 1 client-- No. 2 calculate nodes of establishment virtual machine " is stored in control by load-balanced server
In the calculate node table (compute_nodes tables) of any one database in node, to facilitate the management to user right,
Such as, No. 1 client sends to control node, deletes the request of the virtual machine that No. 2 calculate nodes are created, now control node root
According to the information preserved in database, obtain No. 1 client and this virtual machine, therefore control node were not created on No. 2 nodes
No. 1 request for deleting this virtual machine of client will not be responded.
General, to realize the High Availabitity of database, at least two databases in control node, when to a data
In storehouse calculate node table (compute_nodes tables) after preservation information, this information is copied to other data of this control node
It is identical to ensure the data that all databases in control node are preserved in storehouse calculate node table (compute_nodes tables).Example
Such as, the information of " No. 1 client-- No. 2 calculate nodes of establishment virtual machine " has been stored in a data in control node above
In the calculate node table (compute_nodes tables) in storehouse, after the completion of preservation, this information is copied into other in control node
In the calculate node table (compute_nodes tables) of database, visitor is all preserved with all databases ensured in control node
Family end identical partial data.Even if when one of database breaks down, other databases can be with the number of taking over fault
It is normal that service is externally provided according to the task in storehouse, the High Availabitity of database is realized, and then cause that control node normally externally provides clothes
Business.
By step S102, client is determined a target computing nodes to respond the treatment request of its virtual machine, then
Into step S103.
Step S103:Virtual machine treatment request is sent to the target computing nodes, so that the target is calculated
Node performs the request of data.
In step s 102, load-balanced server determines target computing nodes to respond this client for client
After virtual machine treatment request, load-balanced server sends into control node the virtual machine treatment request of this client
Nova-scheduler, nova-scheduler send to target computing nodes, mesh the virtual machine treatment request of this client
Mark calculate node performs corresponding request of data, for example, create an example, modification one example or deletion in calculate node
One example etc..
If it should be noted that the control node in above-mentioned steps S101 receives the virtual of different clients transmission simultaneously
During machine treatment request, on the one hand, control node can be according to the sequencing for receiving virtual machine treatment request, first, according to step
Rapid S102 processes request to the virtual machine of each client respectively, determines a target computing nodes;Then, in sequence, divide
The virtual machine treatment request of client is not sent to target computing nodes corresponding with each client;Finally, target is calculated
Node performs corresponding action according to the virtual machine treatment request of client.On the other hand, control node can first, to first
Individual client determines target computing nodes, and the virtual machine treatment request of this client then is sent into corresponding target calculates section
Point;Then, target computing nodes are determined to second client, and the virtual machine treatment request of this client is sent to correspondence
Target computing nodes on;Finally, according to the sequencing for receiving virtual machine treatment request, circulate successively, until being last
Individual client determines target computing nodes, and the virtual machine treatment request of this client is sent into corresponding target computing nodes
On.
The embodiment of the present invention 1, is received at the virtual machine of different clients by the load-balanced server in control node
Reason request, load-balanced server monitors the resource service condition of the normal operation calculate node for obtaining according to pacemaker, is
The target computing nodes of each client assignment response its virtual machine treatment request, and by nova-scheduler by this client
The virtual machine treatment request at end is sent to target computing nodes so that target computing nodes perform the virtual machine treatment of this client
Request.
In the process, by the running status of virtual machine in each calculate node in pacemaker monitoring control systems,
The virtual machine of failure in calculate node is isolated, and the virtual machine that will isolate in calculate node to be transferred to other normal
In the calculate node of operation, the High Availabitity of calculate node in OpenStack control systems is realized, and then ensure that to using use
The service at family is uninterrupted.Additionally, in the pacemaker calculate nodes of each normal operation that obtain monitor in real time, being currently running
The quantity of virtual machine and the content of virtual machine operation update the calculate node table (compute_ of any one database
Nodes tables), to realize the management to client jurisdiction.Additionally, load-balanced server is by each client, each client
The virtual machine treatment request at end and the target computing nodes for the distribution of each client, are all stored in control section correspondingly
In the calculate node table (compute_nodes tables) of any one database of point, at least there is two generally in control node
Individual independent database, after information is stored in a database calculate node table (compute_nodes tables), by this information
In copying to other database calculate node tables (compute_nodes tables), and then ensure each database in control node
Identical data are all preserved, when a database breaks down, other databases can work on so that control section
Point incessantly be client service, it is achieved thereby that in a control node database High Availabitity.
Above-described embodiment 1 is described, and database is realized in any one control node in OpenStack control systems
The method of High Availabitity, and while realizing database High Availabitity in a control node, realize OpenStack controls system
The High Availabitity of calculate node in system, final purpose of the invention is to realize OpenStack in OpenStack control systems
High Availabitity, the High Availabitity of OpenStack includes the High Availabitity of control node in OpenStack control systems and calculate node
High Availabitity, has been realized in the High Availabitity of calculate node in OpenStack control systems, for control node in embodiment 1
High Availabitity, i.e., when in OpenStack control systems control node break down when, OpenStack control systems other
Control node, the task of the adapter now control node of failure so that OpenStack control systems customer in response end incessantly
The virtual machine treatment request for sending.The High Availabitity implementation process of control node is described in detail by embodiment 2.
Embodiment 2
The embodiment of the present invention is discussed in detail a kind of method for realizing control node High Availabitity in OpenStack control systems,
The method is applied to the control system based on OpenStack, in the present embodiment OpenStack control systems include scheduler,
Two control nodes for similarly configuring and Duo Tai calculate nodes.
Fig. 2, is a kind of method flow diagram for realizing OpenStack High Availabitities provided in an embodiment of the present invention, specific steps
Including:
Step S200:Client sends virtual machine treatment request to scheduler.
The embodiment of the present invention, during the High Availabitity of control node in realizing OpenStack control systems,
Two control nodes in OpenStack control systems can be the virtual machine treatment request that client sends, and distribute target
Calculate node, target computing nodes are distributed for the virtual machine treatment request that client sends by which platform control node, by
Load-balanced server in control node is responsible for, and a load balancing service is all respectively present in two control nodes
Device, wherein for being responsible for the load-balanced server that client distributes target control node, referred to as scheduler.So, adjust
Degree device, geographically, is located in control node, is same with control node and calculate node in logic function
Level.Therefore, the treatment request of its virtual machine is sent to scheduler by client so that scheduler is its distribution target control section
Point, and then cause the target computing nodes that target control node is its request of its assignment response.
Step S201:Scheduler receives the virtual machine treatment request that client sends.
Scheduler receives the virtual machine treatment request from client, and distribution target control node is asked for this.
Step S202:The loading condition of scheduler each control node described in, it is determined that responding the virtual machine treatment
One target control node of request.
In this step, scheduler is obtained from the internal memory of each control node of OpenStack control systems, each control section
Loading condition of the point in each calculate node of control system, target control is defined as by the control node corresponding to minimal negative carrying capacity
Node processed, specifically includes step B1~step B2.
Step B1:Scheduler obtains the loading condition of each control node, described each control node from internal memory
Loading condition represent, each control node just customer in response end virtual machine treatment request quantity.
In each control node in OpenStack control systems, at the client virtual machine that is currently responding
Reason request, is all stored in the internal memory of each control node in the way of a kind of interim storage, and the mode of the interim storage refers to,
After the virtual machine treatment request of client is responded, the internal memory shared by client virtual machine treatment that this part is responded is empty
Between be released, the memory headroom being released processes request for storing virtual machine transmitted by later client, circulates successively, makes
Each control node memory headroom stored be all each control node responding client virtual machine treatment request, this
Planting the mode of interim storage can improve the internal memory speed of service of each control node.
First, that the client that each control node currently responding is obtained from the internal memory of each control node is virtual for scheduler
Machine treatment request;Then, the client virtual machine treatment request that each control node is being responded is counted, each control node is obtained
Loading condition.
For example, the control node of OpenStack control systems is respectively A control nodes and B control nodes, A control nodes
In load-balanced server be scheduler, scheduler is obtained from the internal memory of A and B control nodes respectively, A control nodes and B
The active client virtual machine treatment request of control node, and count the active client for obtaining A control nodes and B control nodes
Virtual machine processes request amount.Such as, scheduler statistics obtains having 7 client virtual machines for responding in A control nodes
Treatment request, have in B control nodes 5 responding client virtual machines treatment request, therefore, A control nodes it is current
Load capacity is that the current load capacity of 7, B control nodes is 5.
Step B2:Scheduler according to each control node loading condition, by the control section corresponding to minimal negative carrying capacity
Point, is defined as responding the target control node of the virtual machine treatment request.
Scheduler has obtained the loading condition of each control node in step bl is determined., and scheduler is by corresponding to minimal negative carrying capacity
Control node be defined as target control node, this target control node is the void that this client assignment response this client sends
The target computing nodes of plan machine treatment request.For the example in step B1, the load capacity that scheduler has obtained A control nodes is
The load capacity of 7, B control nodes is 5, because the load capacity of B control nodes is small, so, B control nodes are defined as mesh by scheduler
Mark control node.
Step S203:Virtual machine treatment request is sent to the target control node by scheduler.
After scheduler determines target control node, the client virtual machine treatment request of reception is sent to target control
Node, to facilitate the target computing nodes of this request of target control node assignment response.For the example of step S202, scheduler
B control nodes are had determined that for target control node, in this step, virtual machine treatment request is sent to B control sections by scheduler
Point.
Step S204:Target control node receives the virtual machine treatment request of the scheduler forwarding, according to default negative
The target computing nodes that equalization algorithm determines the response virtual machine treatment request are carried, and virtual machine treatment request is sent
To the target computing nodes.
In this step, first, target control node receives the virtual machine treatment request of scheduler forwarding;Then, target
The method of load-balanced server in control node according to step S102 in embodiment 1, it is determined that responding the virtual machine
Process the target computing nodes of request.
For the example in step S203, virtual machine treatment request is sent to B control nodes by scheduler, in this step
In rapid, B control nodes are according to the method described in step S102 in embodiment 1, it is assumed that now, and the resource of No. 3 calculate nodes is used
At least, then No. 3 calculate nodes are defined as target computing nodes by B control nodes, and will be virtual by nova-scheduler
Machine treatment request is sent to No. 3 calculate nodes.
It should be noted that after this step is performed, target control node will now " Client-client be specifically empty
Plan machine treatment request-target computing nodes " information is stored in the calculate node table of any one database of target control node
In (compute_nodes tables), and this information is copied to other databases of control node in OpenStack control systems
Calculate node table (compute_nodes tables) in so that the data that all data places of OpenStack control systems are preserved
It is all identical.
Step S205:Target computing nodes perform the request of data.
After target computing nodes receive the virtual machine treatment request of nova-scheduler transmissions in target control node,
Recognize that this is asked first, and prepared to perform this request, it is to be prepared after the completion of, perform this request of data.
The embodiment of the present invention, first, scheduler receives the virtual machine treatment request that client sends, and obtains control node
The client virtual machine treatment request responded in internal memory, the client that each control node that statistics is obtained is being responded is virtual
The amount of machine treatment request, obtains the loading condition of each control node;Then, scheduler is according to the loading condition of each control node,
The small control node of load capacity is defined as target control node, and the treatment request of this virtual machine is sent to this target control section
Point;Finally, this target control node determines the small calculate node of load capacity according to the load capacity of each normal operation calculate node
It is target computing nodes so that target computing nodes perform this request of data, additionally, control node will " Client-client has
The virtual machine treatment request-target computing nodes of body " are stored in the calculate node of all databases of OpenStack control systems
In table (compute_nodes tables) so that the data of database are all identical, when a control section of OpenStack control systems
When point breaks down, another control node can be with the task of taking over fault control node so that for the service that client provides
Uninterruptedly, so as to realize the High Availabitity of control node in OpenStack control systems.
Embodiment 3
The embodiment of the invention discloses a kind of device for realizing data High Availabitity, this device is integrated in based on OpenStack
Control system any one control node on, the control system includes two control nodes and multiple calculate nodes;Please
Referring to Fig. 3, described device includes:
Request unit 301 is received, the virtual machine for receiving different clients transmission processes request;
Target computing nodes unit 302 is determined, for determining the response virtual machine according to default load-balancing algorithm
Process the target computing nodes of request;
Request unit 303 is sent, for virtual machine treatment request to be sent to the target computing nodes, with toilet
State target computing nodes and perform the request of data.
The embodiment of the invention discloses a kind of device for realizing data High Availabitity, also include:
Virtual machine processing information unit is obtained, for obtaining the virtual machine processing information in the virtual machine treatment request,
The virtual machine processing information includes:Virtual machine creating information, deletion information or modification information;
Storage virtual machine processing information unit, for by the client and the corresponding virtual machine processing information
Store into the calculate node table of database.
Monitoring unit, the running status for monitoring the virtual machine in the multiple calculate node;
Statistic unit, the resource service condition of each calculate node is counted for the running status according to the virtual machine;
Storage unit, the calculate node table for the resource service condition of each calculate node to be stored in database
In.
Preferably, the storage virtual machine processing information unit, specifically includes:
Memory module, for the client to be stored to any one database with corresponding virtual machine processing information
In calculate node table;
Replication module, the calculating for the client to be copied to other databases with corresponding virtual machine processing information
In node table.
Preferably, the determination target computing nodes unit, specifically for resource is determined using minimum calculate node
To respond the target computing nodes of the virtual machine treatment request.
The embodiment of the present invention, first, by receiving unit, receives the virtual machine treatment request that different clients send;So
Afterwards, by obtaining virtual machine processing information unit, monitoring unit, statistic unit, storage unit and determining destination node unit,
It is the virtual machine treatment request distribution target computing nodes that client sends, and by storage virtual machine processing information unit, will
The client and the corresponding virtual machine processing information are stored into the calculate node table of database;Finally, pass through
Request unit is sent, virtual machine treatment request is sent to the target computing nodes, so as to the target computing nodes
Perform the request of data.
Embodiment 4
The embodiment of the invention discloses a kind of system for realizing OpenStack High Availabitities, the system integration is in being based on
The control system of OpenStack, the control system based on OpenStack includes:Scheduler, two configuration identical controls
Node and multiple calculate nodes, the scheduler include receiving unit, determine target control node unit and transmitting element, please join
See Fig. 4, including:
Receiving unit 401, the virtual machine for receiving client transmission processes request;
Target control node unit 402 is determined, for the loading condition according to each control node, it is determined that response institute
State the target control node of virtual machine treatment request;
Transmitting element 403, for virtual machine treatment request to be sent into the target control node;
Wherein, the device in the control node described in Integration Data Model example 3;
The calculate node is used to perform the request of data.
Preferably, the determination target control node unit, including:
Loading condition module is obtained, the load feelings for obtaining each control node from each control node internal memory
Condition;
Target control node module is determined, for the loading condition according to each control node, by minimal negative carrying capacity
Corresponding control node, is defined as responding the target control node of the virtual machine treatment request.
The embodiment of the present invention, first, by receiving unit, receives the virtual machine treatment request that client sends;Then, lead to
Cross in determination target control node unit, obtain load table module with target control node module is determined to determine to respond institute
State the target control node of virtual machine treatment request;Finally, by transmitting element, virtual machine treatment request is sent to institute
State target control node.When a control node in control system breaks down, the control node of other normal works connects
The task of pipe malfunctioning node, is the virtual machine treatment request distribution target computing nodes of client, and then realizes OpenStack's
The High Availabitity of control node in control system.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference with other embodiment, between each embodiment identical similar part mutually referring to.
For device class embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, related part ginseng
See the part explanation of embodiment of the method.
Finally, in addition it is also necessary to explanation, herein, such as first and second or the like relational terms be used merely to by
One entity or operation make a distinction with another entity or operation, and not necessarily require or imply these entities or operation
Between there is any this actual relation or order.And, term " including ", "comprising" or its any other variant meaning
Covering including for nonexcludability, so that process, method, article or equipment including a series of key elements not only include that
A little key elements, but also other key elements including being not expressly set out, or also include for this process, method, article or
The intrinsic key element of equipment.In the absence of more restrictions, the key element limited by sentence "including a ...", does not arrange
Except also there is other identical element in the process including the key element, method, article or equipment.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or uses the present invention.
Various modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, the present invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The scope most wide for causing.
Claims (14)
1. a kind of method for realizing database High Availabitity, it is characterised in that methods described is applied to the control based on OpenStack
In any one control node of system, the control system includes two control nodes and multiple calculate nodes;The method bag
Include:
Receive the virtual machine treatment request that different clients send;
The target computing nodes of the response virtual machine treatment request are determined according to default load-balancing algorithm;
Virtual machine treatment request is sent to the target computing nodes, so that the target computing nodes perform the number
According to request.
2. method according to claim 1, it is characterised in that also include:
The virtual machine processing information in the virtual machine treatment request is obtained, the virtual machine processing information includes:Virtual machine is created
Build information, deletion information or modification information;
The client and the corresponding virtual machine processing information are stored into the calculate node table of database.
3. method according to claim 2, it is characterised in that the control node includes at least two numbers independent mutually
It is according to storehouse, then described to store into database the virtual machine creating information, deletion information or modification information, including:
The control node stores to the calculating of any one database the client with corresponding virtual machine processing information
In node table;
The client is copied to the control node calculate node of other databases with corresponding virtual machine processing information
In table.
4. method according to claim 1, it is characterised in that also include:
Monitor the running status of the virtual machine in the multiple calculate node;
The resource service condition of each calculate node is counted according to the running status of the virtual machine;
The resource service condition of each calculate node is stored in the calculate node table of database.
5. method according to claim 4, it is characterised in that described to determine response institute according to default load-balancing algorithm
The target computing nodes of virtual machine treatment request are stated, including:
Resource is defined as using minimum calculate node to respond the target computing nodes that the virtual machine processes request.
6. a kind of method for realizing OpenStack High Availabitities, it is characterised in that methods described is applied to based on OpenStack
Control system, the control system based on OpenStack includes:Scheduler, two configurations identical control node and multiples
Calculate node, the method includes:
The scheduler receives the virtual machine treatment request that client sends;
The loading condition of the scheduler each control node described in, it is determined that responding of the virtual machine treatment request
Target control node;
Virtual machine treatment request is sent to the target control node by the scheduler;
The target control node receives the virtual machine treatment request of the scheduler dispatches, according to default load-balancing algorithm
It is determined that responding the target computing nodes of the virtual machine treatment request, and virtual machine treatment request is sent to the target
Calculate node;
The target computing nodes perform the request of data.
7. the method according to the claim 6, it is characterised in that the scheduler is according to each control node
Loading condition, it is determined that the target control node of the virtual machine treatment request is responded, including:
The scheduler obtains the loading condition of each control node, the load feelings of each control node from internal memory
Condition is stored in the respective internal memory of the control node;
The scheduler according to each control node loading condition, by the control node corresponding to minimal negative carrying capacity, really
It is set to the target control node of the response virtual machine treatment request.
8. a kind of device for realizing data High Availabitity, it is characterised in that described device is integrated in the control system based on OpenStack
In any one control node of system, the control system includes two control nodes and multiple calculate nodes;Described device bag
Include:
Request unit is received, the virtual machine for receiving different clients transmission processes request;
Target computing nodes unit is determined, for determining the response virtual machine treatment request according to default load-balancing algorithm
Target computing nodes;
Request unit is sent, for virtual machine treatment request to be sent to the target computing nodes, so as to the target
Calculate node performs the request of data.
9. device according to claim 8, it is characterised in that described device also includes:
Virtual machine processing information unit is obtained, it is described for obtaining the virtual machine processing information in the virtual machine treatment request
Virtual machine processing information includes:Virtual machine creating information, deletion information or modification information;
Storage virtual machine processing information unit, for the client and the corresponding virtual machine processing information to be stored
Into the calculate node table of database.
10. device according to claim 9, it is characterised in that the control node includes that at least two is independent mutually
Database, the storage virtual machine processing information unit includes:
Memory module, for the client to be stored to the calculating of any one database with corresponding virtual machine processing information
In node table;
Replication module, the calculate node for the client to be copied to other databases with corresponding virtual machine processing information
In table.
11. devices according to claim 8, it is characterised in that described device also includes:
Monitoring unit, the running status for monitoring the virtual machine in the multiple calculate node;
Statistic unit, the resource service condition of each calculate node is counted for the running status according to the virtual machine;
Storage unit, for the resource service condition of each calculate node to be stored in the calculate node table of database.
12. devices according to claim 11, it is characterised in that the determination target computing nodes unit, specifically for
Resource is defined as using minimum calculate node to respond the target computing nodes that the virtual machine processes request.
13. a kind of systems for realizing OpenStack High Availabitities, it is characterised in that the system integration is in based on OpenStack's
Control system, the control system based on OpenStack includes:Scheduler, two configurations identical control node and multiples
Calculate node, the scheduler includes receiving unit, determines target control node unit and transmitting element, and the receiving unit is used
Request is processed in the virtual machine for receiving client transmission;The determination target control node unit, for according to described each control
The loading condition of node processed, it is determined that responding the target control node of the virtual machine treatment request;The transmitting element, for inciting somebody to action
The virtual machine treatment request is sent to the target control node;
The integrated device as described in claim 8~12 any one in the control node;
The calculate node is used to perform the request of data.
14. systems according to claim 13, it is characterised in that the determination target control node unit, including:
Loading condition module is obtained, the loading condition for obtaining each control node from each control node internal memory;
Determine target control node module, it is for the loading condition according to each control node, minimal negative carrying capacity institute is right
The control node answered, is defined as responding the target control node of the virtual machine treatment request.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611265200.5A CN106775953A (en) | 2016-12-30 | 2016-12-30 | Realize the method and system of OpenStack High Availabitities |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611265200.5A CN106775953A (en) | 2016-12-30 | 2016-12-30 | Realize the method and system of OpenStack High Availabitities |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106775953A true CN106775953A (en) | 2017-05-31 |
Family
ID=58951712
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611265200.5A Pending CN106775953A (en) | 2016-12-30 | 2016-12-30 | Realize the method and system of OpenStack High Availabitities |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106775953A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107222352A (en) * | 2017-06-30 | 2017-09-29 | 郑州云海信息技术有限公司 | One kind management network recovery method and device |
CN107239383A (en) * | 2017-06-28 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of failure monitoring method and device of OpenStack virtual machines |
CN108089911A (en) * | 2017-12-14 | 2018-05-29 | 郑州云海信息技术有限公司 | The control method and device of calculate node in OpenStack environment |
CN108234645A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of task |
CN110580198A (en) * | 2019-08-29 | 2019-12-17 | 上海仪电(集团)有限公司中央研究院 | Method and device for adaptively switching OpenStack computing node into control node |
CN111897626A (en) * | 2020-07-07 | 2020-11-06 | 烽火通信科技股份有限公司 | Cloud computing scene-oriented virtual machine high-reliability system and implementation method |
CN112199176A (en) * | 2020-10-16 | 2021-01-08 | 济南浪潮数据技术有限公司 | Service processing method, device and related equipment |
TWI764160B (en) * | 2020-06-01 | 2022-05-11 | 鴻海精密工業股份有限公司 | An implementation method, a device for high availability based on openstack, and an electronic device |
US11388274B2 (en) | 2020-06-01 | 2022-07-12 | Hon Hai Precision Industry Co., Ltd. | Method for implementing high availability of bare metal node based on OpenStack and electronic device using the same |
WO2023065922A1 (en) * | 2021-10-22 | 2023-04-27 | 中移(苏州)软件技术有限公司 | Interactive method, computer device, and computer storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102479099A (en) * | 2010-11-22 | 2012-05-30 | 中兴通讯股份有限公司 | Virtual machine management system and use method thereof |
WO2016165304A1 (en) * | 2015-04-16 | 2016-10-20 | 华为技术有限公司 | Method for managing instance node and management device |
CN106254471A (en) * | 2016-08-09 | 2016-12-21 | 华为技术有限公司 | Resource United Dispatching method and system under a kind of isomery cloud environment |
-
2016
- 2016-12-30 CN CN201611265200.5A patent/CN106775953A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102479099A (en) * | 2010-11-22 | 2012-05-30 | 中兴通讯股份有限公司 | Virtual machine management system and use method thereof |
WO2016165304A1 (en) * | 2015-04-16 | 2016-10-20 | 华为技术有限公司 | Method for managing instance node and management device |
CN106254471A (en) * | 2016-08-09 | 2016-12-21 | 华为技术有限公司 | Resource United Dispatching method and system under a kind of isomery cloud environment |
Non-Patent Citations (1)
Title |
---|
李兴等: "基于OpenStack 架构上的高可用实现", 《电脑编程技巧与维护》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107239383A (en) * | 2017-06-28 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of failure monitoring method and device of OpenStack virtual machines |
CN107222352A (en) * | 2017-06-30 | 2017-09-29 | 郑州云海信息技术有限公司 | One kind management network recovery method and device |
CN107222352B (en) * | 2017-06-30 | 2020-09-18 | 苏州浪潮智能科技有限公司 | Management network recovery method and device |
CN108089911A (en) * | 2017-12-14 | 2018-05-29 | 郑州云海信息技术有限公司 | The control method and device of calculate node in OpenStack environment |
CN108234645A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of task |
CN110580198A (en) * | 2019-08-29 | 2019-12-17 | 上海仪电(集团)有限公司中央研究院 | Method and device for adaptively switching OpenStack computing node into control node |
TWI764160B (en) * | 2020-06-01 | 2022-05-11 | 鴻海精密工業股份有限公司 | An implementation method, a device for high availability based on openstack, and an electronic device |
US11388274B2 (en) | 2020-06-01 | 2022-07-12 | Hon Hai Precision Industry Co., Ltd. | Method for implementing high availability of bare metal node based on OpenStack and electronic device using the same |
CN111897626A (en) * | 2020-07-07 | 2020-11-06 | 烽火通信科技股份有限公司 | Cloud computing scene-oriented virtual machine high-reliability system and implementation method |
CN112199176A (en) * | 2020-10-16 | 2021-01-08 | 济南浪潮数据技术有限公司 | Service processing method, device and related equipment |
CN112199176B (en) * | 2020-10-16 | 2023-01-17 | 济南浪潮数据技术有限公司 | Service processing method, device and related equipment |
WO2023065922A1 (en) * | 2021-10-22 | 2023-04-27 | 中移(苏州)软件技术有限公司 | Interactive method, computer device, and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106775953A (en) | Realize the method and system of OpenStack High Availabitities | |
CN109451065B (en) | Soft load balancing and shunting automation system and operation method thereof | |
US10122595B2 (en) | System and method for supporting service level quorum in a data grid cluster | |
CN106663030B (en) | Scalable failover communication in distributed clusters | |
CN105245373B (en) | A kind of container cloud platform system is built and operation method | |
CN106899518A (en) | A kind of method for processing resource and device based on Internet data center | |
CN105794182B (en) | The processing method and its system of server failure are locked in distributed system | |
CN104463492A (en) | Operation management method of electric power system cloud simulation platform | |
US20060155912A1 (en) | Server cluster having a virtual server | |
CN107925633A (en) | Data center resource tracks | |
CN108270726B (en) | Application instance deployment method and device | |
CN103888277B (en) | A kind of gateway disaster-tolerant backup method, device and system | |
CN101984632A (en) | Load distributing method, device and server in distributed cache system | |
EP3180692A1 (en) | Fault tolerant federation of computing clusters | |
CN108259175B (en) | Distributed password service method and system | |
CN109845192B (en) | Computer system and method for dynamically adapting a network and computer readable medium | |
CN105630589A (en) | Distributed process scheduling system and process scheduling and execution method | |
US10331472B2 (en) | Virtual machine service availability | |
CN106936623A (en) | The management method of distributed cache system and cache cluster | |
CN103458055A (en) | Clout competing platform | |
CN106603696A (en) | High-availability system based on hyper-converged infrastructure | |
CN106980529A (en) | The computer system of baseboard management controller resource management | |
CN110445662A (en) | OpenStack control node is adaptively switched to the method and device of calculate node | |
JP2012231636A (en) | Monitoring control system for electric power system | |
CN107197002A (en) | Cloud computing system and cloud data processing method |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
RJ01 | Rejection of invention patent application after publication |