CN109951531A - Super fusion cloud computing system - Google Patents
Super fusion cloud computing system Download PDFInfo
- Publication number
- CN109951531A CN109951531A CN201910145374.5A CN201910145374A CN109951531A CN 109951531 A CN109951531 A CN 109951531A CN 201910145374 A CN201910145374 A CN 201910145374A CN 109951531 A CN109951531 A CN 109951531A
- Authority
- CN
- China
- Prior art keywords
- fusion
- super
- machine
- module
- memory module
- 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
Links
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to fields of communication technology, provide a kind of super fusion cloud computing system, including switch, which is characterized in that further include at least one super aggregators, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion all-in-one machine is communicatively coupled by the switch;Each super fusion all-in-one machine includes computing module, network module, memory module and control module;The computing module obtains corresponding resource occupation information for analyzing the resource service condition in the computing module and memory module;The resource service condition includes the operation utilization rate of computing module and the storage occupancy of memory module;The control module is used to determine whether the super fusion all-in-one machine needs to carry out expanding storage depth according to the resource occupation information, and passes sequentially through the network module and the switch when needing to carry out expanding storage depth and externally carry out longitudinal capacity extension or lateral capacity extension.
Description
Technical field
The invention belongs to fields of communication technology, more particularly to a kind of super fusion cloud computing system.
Background technique
It was known as the mainstream of data center between the more than ten years in past based on the data center architecture of virtualization technology, but it is still
It is that the traditional three-tier architecture to be formed is superimposed with server, storage network, the storage big component of equipment three, as shown in Figure 1.With application
With the surge of number of users, the three-tier architecture in Fig. 1 exposes problems during Constructing data center and O&M, basic former
Because being that the performance of the shared storage of centralization can not really realize on-demand dilatation with business development.Storage control determines storage
Overall performance, because be embedded within storage equipment in, cannot achieve extending transversely.For virtual machine quantity and I/O
When the demand of (Input/Output, input/output) interface increases significantly, the data access request of flood tide will lead to controller and gather around
Plug, causes performance bottleneck.It therefore, can only be existing by purchasing new storage equipment solution when there is new application to need online
Storage device performance bottleneck problem in environment.
Lack linear expansion ability just because of storage equipment, cause business development limited, business hair can not be timely responded to
Exhibition demand.Store advanced feature such as data compression, data de-duplication, duplication characteristic and SAN (optical fiber) interchanger of equipment
Port be both needed to the authorization outside purchase volume and could activate use, it means that the increase of cost.
In conclusion the growth and user with number of applications and userbase are mentioned for what system availability required
Height, three-tier architecture can not adapt to current enterprises and institutions' industry because of the deficiencies of its framework is complicated, maintenance difficulties are big, at high cost
It is engaged in the requirement of development, Constructing data center and administrative department need a kind of new architecture to solve above-mentioned technical problem.
Summary of the invention
In view of this, the embodiment of the invention provides the super fusion cloud computing systems of one kind, to solve three layers in the prior art
Framework is because of the technical problem that its framework is complicated, maintenance difficulties are big and at high cost.
The embodiment of the invention provides the super fusion cloud computing systems of one kind comprising switch, and further include
At least one super aggregators, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion one
Machine is communicatively coupled by the switch.
Each super fusion all-in-one machine includes computing module, network module, memory module and control module.
The computing module is obtained for analyzing the resource service condition in the computing module and memory module
Corresponding resource occupation information;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module accounts for
With rate.
The control module is used to determine whether the super fusion all-in-one machine needs to carry out according to the resource occupation information
Expanding storage depth, and the network module and the switch are passed sequentially through when needing to carry out expanding storage depth
Externally carry out longitudinal capacity extension or lateral capacity extension.
The longitudinal direction capacity extension are as follows: the data backup carried out between multiple super fusion all-in-one machines in same super aggregators
And/or Data Migration.
The transverse direction capacity extension are as follows: when the quantity of the super aggregators is greater than one, in different super aggregators
Between the Data Migration that is carried out.
The super fusion cloud computing system provided in the embodiment of the present invention, including switch and at least one super fusion
Node, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion all-in-one machine passes through the data
Switching equipment is communicatively coupled;Each super fusion all-in-one machine includes computing module, network module, memory module and control
Module;The computing module obtains phase for analyzing the resource service condition in the computing module and memory module
The resource occupation information answered;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module occupies
Rate;The control module is used to determine whether the super fusion all-in-one machine needs to carry out storage appearance according to the resource occupation information
Amount extension, and pass sequentially through when needing to carry out expanding storage depth the network module and the switch externally into
Row longitudinal direction capacity extension or lateral capacity extension;The longitudinal direction capacity extension are as follows: multiple super fusions one in same super aggregators
The data backup carried out between body machine and/or Data Migration;The transverse direction capacity extension are as follows: in the quantity of the super aggregators
When greater than one, the Data Migration that is carried out between different super aggregators.Super fusion cloud computing in the embodiment of the present invention
Computing module, network module, memory module and control module are arranged in the same super fusion all-in-one machine framework of system, i.e.,
On the same physical node, framework is simple, reduces hardware cost, while can calculate according in current super aggregators
The resource occupation information of module and memory module neatly carries out extending transversely or Longitudinal Extension, solves memory module performance
The problem of bottleneck, has stronger practicability and ease for use.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the schematic diagram of traditional three-tier architecture provided in an embodiment of the present invention;
Fig. 2 is the schematic diagram for the super fusion cloud computing system that the embodiment of the present invention one provides;
Fig. 3 is the structural schematic diagram for the super aggregators that the embodiment of the present invention one provides;
Fig. 4 is the structural schematic diagram for the super fusion all-in-one machine that the embodiment of the present invention one provides;
Fig. 5 is the schematic diagram of super fusion cloud computing system provided by Embodiment 2 of the present invention.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate technical solution of the present invention, the following is a description of specific embodiments.
Embodiment one:
Fig. 2 shows the schematic diagram for the super fusion cloud computing system that the embodiment of the present invention one provides, Fig. 3 shows the present invention
The structural schematic diagram for the super aggregators that embodiment one provides;Fig. 4 shows the super fusion one of the offer of the embodiment of the present invention one
The structural schematic diagram of machine.With reference to Fig. 2, Fig. 3 and Fig. 4, which includes that switch and two super are melted
Node is closed, each super aggregators include three super fusion all-in-one machines;Described three super fusion all-in-one machines pass through the number
It is communicatively coupled according to switching equipment;Each super fusion all-in-one machine includes computing module, network module, memory module and control
Molding block;The computing module is obtained for analyzing the resource service condition in the computing module and memory module
Corresponding resource occupation information;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module accounts for
With rate;The control module is used to determine whether the super fusion all-in-one machine is stored according to the resource occupation information
Capacity extension, and the network module is passed sequentially through when needing to carry out expanding storage depth and the switch is external
Carry out longitudinal capacity extension or lateral capacity extension.
Wherein, longitudinal capacity extension are as follows: the number carried out between multiple super fusion all-in-one machines in same super aggregators
According to backup and/or Data Migration;The transverse direction capacity extension are as follows: when the quantity of the super aggregators is greater than one, not
With the Data Migration carried out between super aggregators.
Optionally, the computing module is for dividing the resource service condition in the computing module and memory module
Analysis, obtains corresponding resource occupation information, specifically includes:
Computing module is for determining resource comprehensive to the operation utilization rate of computing module and the storage occupancy of memory module
Occupancy, using resource comprehensive occupancy as corresponding resource occupation information.
It further, can be according to the needs of analysis, respectively the storage occupancy setting of operation utilization rate and memory module
Corresponding weight.For example, the weight of operation utilization rate is W1, the storage occupancy W2 of memory module, then resource comprehensive occupancy
=operation utilization rate * W1+ stores occupancy * W2.For example, operation utilization rate is 60% (i.e. 0.6), the weight of operation utilization rate is
W1=0.6, storage occupancy are 70% (i.e. 0.7), store occupancy W2=0.4, then resource comprehensive occupancy=0.6*0.6+
0.7*0.4=0.64.
Optionally, the control module be used for according to the resource occupation information determine the super fusion all-in-one machine whether need
Expanding storage depth is carried out, and passes sequentially through the network module and data friendship when needing to carry out expanding storage depth
Exchange device externally carries out longitudinal capacity extension or lateral capacity extension, specifically:
It include the first surpassing fusion all-in-one machine the second to surpass fusion all-in-one machine at least one in the same super aggregators;Institute
It states and the first surpasses in fusion all-in-one machine, control module is true according to the operation utilization rate of computing module and the storage occupancy of memory module
Determine resource comprehensive occupancy;When the resource comprehensive occupancy is greater than predetermined occupancy, control module passes through network module, institute
State switch and the network module for the second surpassing fusion all-in-one machine with it is described the second surpass merge the control mould of all-in-one machine
Block is communicated, when determining that the memory module for the second surpassing fusion all-in-one machine meets Longitudinal Extension condition, described first
In super fusion all-in-one machine, control module control memory module the second is surpassed by network module and the switch to described
It merges all-in-one machine and carries out Longitudinal Extension.
When it is described the second surpass fusion all-in-one machine memory module it is impossible to meet when Longitudinal Extension demand, the first surpass described
It merges in all-in-one machine, control module control memory module passes sequentially through network module and the switch and melts to outside is super
It is extending transversely to close node progress.
For example, super fusion cloud computing system includes two super aggregators, each super aggregators include 3 super fusions one
Body machine, the memory module of the super fusion all-in-one machine in each super aggregators is it is impossible to meet when Longitudinal Extension demand, then
Determine that the configuration of the memory module of the super fusion all-in-one machine in current super aggregators is unable to satisfy Longitudinal Extension demand.If
The memory module of super fusion all-in-one machine in one of them super aggregators it is impossible to meet when Longitudinal Extension demand, then judges
When whether the memory module of the super fusion all-in-one machine in another super aggregators can satisfy Longitudinal Extension demand, if another
It is impossible to meet Longitudinal Extension demands for the memory module of super fusion all-in-one machine in a super aggregators, then one super
Aggregators the first surpass in fusion all-in-one machine, and control module control memory module passes sequentially through network module and the data are handed over
Exchange device carries out extending transversely to external super aggregators.It is understood that if super fusion in another super aggregators
The memory module of all-in-one machine can satisfy Longitudinal Extension demand, then carries out Longitudinal Extension.
Wherein, longitudinal capacity extension specifically:
By the Data Migration in the memory module for the first surpassing fusion all-in-one machine to the memory module for the second surpassing fusion all-in-one machine
It is backed up, after the memory module capacity amplification for the first surpassing fusion all-in-one machine, then will backup to and described the second surpass fusion
The Data Migration of the memory module of all-in-one machine to it is described the first surpass fusion all-in-one machine in capacity amplification after memory module;It is same
It include described the first surpassing fusion all-in-one machine and at least one described the second surpasses fusion all-in-one machine in super aggregators.
Optionally, the migration is online thermophoresis.Wherein, the amplification of capacity includes but is not limited to CPU core number, memory
2TB, the CPU core after upgrading are inside saved as example, the CPU core number of current super fusion all-in-one machine is 8 cores with the HardwareUpgring of disk
Number is 16 cores, inside saves as 8TB.
Wherein, the lateral capacity extension specifically:
By the Data Migration in the memory module for the first surpassing fusion all-in-one machine to the memory module for the second surpassing fusion all-in-one machine;
It is described the first surpass fusion all-in-one machine and it is described the second surpass fusion all-in-one machine do not belong to a super aggregators.
For example, super fusion cloud computing system includes a super aggregators A, super aggregators A includes 3 super fusion one
Machine a1, a2 and a3 need to increase at least one super aggregators according to extending transversely when needing extending transversely, it is assumed that increase
Adding super aggregators B, super aggregators B includes super fusion all-in-one machine b1, b2 and b3, then will merge all-in-one machine a1, a2 and/or a3
Memory module in Data Migration to it is super fusion all-in-one machine b1, b2 and/or b3 in memory module.It is understood that working as
It, can depositing one fusion all-in-one machine (a1, a2 or a3) in super aggregators A when demand extending transversely is the first demand
Store up module in Data Migration to it is super fusion all-in-one machine b1, b2 or b3 in memory module;When demand extending transversely is second
It, can be by two fusion all-in-one machines (a1, a2 in super aggregators A when demand;A1, a2 or a2, a3) memory module in
Data move to the memory module in super fusion all-in-one machine b1, b2 or b3 together;When demand extending transversely is third demand,
Data in the memory module of super fusion all-in-one machine (a1, a2 and a3) of three in super aggregators A can be moved to together
Memory module in super fusion all-in-one machine b1, b2 or b3;When demand extending transversely is four demand, super fusion can be saved
Two super fusion all-in-one machine (a1, a2 in point A;A1, a3 or a2, a3) memory module in data respectively correspond and move to two
A super fusion all-in-one machine (b1, b2;B1, b3 or b2, b3) in memory module, for example, by the super fusion one in super aggregators A
Data Migration in the memory module of body machine a1 melts super in super aggregators A to the memory module of super fusion all-in-one machine b1
Close memory module of the Data Migration in the memory module of all-in-one machine a2 to super fusion all-in-one machine b2, surpassing in super aggregators A
The data merged in the memory module of all-in-one machine a3 do not migrate.When demand extending transversely is five demand, can will surpass
Data in the memory module of three super fusions all-in-one machine (a1, a2 and a3) in aggregators A, which respectively correspond, moves to three
Memory module in super fusion all-in-one machine (b1, b2 and b3), for example, by the data point in the memory module of super fusion all-in-one machine a1
The memory module in super fusion all-in-one machine b1 Dui Ying not be moved to, the data in the memory module of super fusion all-in-one machine a2 are distinguished
The corresponding memory module moved in super fusion all-in-one machine b2 is right respectively by the data in the memory module of super fusion all-in-one machine a3
The memory module in super fusion all-in-one machine b3 should be moved to.
It should be noted that above-mentioned first demand, the second demand, third demand, the 4th demand and the 5th demand are different
Demand, optionally, the first demand, the second demand, third demand, the 4th demand and the 5th demand are sequentially increased.Optionally, institute
State the access frequency that computing module is also used to calculate data in the memory module.
The data that the control module is also used to by access frequency in the memory module lower than predeterminated frequency pass sequentially through
The network module and the switch are migrated to the memory module of external super fusion all-in-one machine.
Wherein, access frequency refers to the number accessed in preset a period of time.For example, predeterminated frequency is to access in 60 days
2 times, then access frequency (for example, accessing 1 time in 60 days) is lower than predeterminated frequency (accessing 2 times in 60 days) in the memory module
Data need to migrate, and are migrated by the network module and the switch to the storage of external super fusion all-in-one machine
Module.It should be noted that the memory module of external super fusion all-in-one machine refer to relative to current super fusion all-in-one machine it
Outer super fusion all-in-one machine, super fusion all-in-one machine including the same super aggregators and does not belong to the same super aggregators
Super fusion all-in-one machine.
Optionally, the memory module is multiple disks for storing data.
Optionally, the super aggregators include at least three surpassing fusion all-in-one machine.It is understood that each super fusion
Node can also include four super fusion all-in-one machines, or the super aggregators in part include three super fusion all-in-one machines, and part is super
Aggregators include four super fusion all-in-one machines, and this is not restricted.
The super fusion cloud computing system provided in the embodiment of the present invention, including switch and multiple super fusion sections
Point, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion all-in-one machine is handed over by the data
Exchange device is communicatively coupled;Each super fusion all-in-one machine includes computing module, network module, memory module and control mould
Block;The computing module obtains corresponding for analyzing the resource service condition in the computing module and memory module
Resource occupation information;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module occupies
Rate;The control module is used to determine whether the super fusion all-in-one machine needs to carry out storage appearance according to the resource occupation information
Amount extension, and pass sequentially through when needing to carry out expanding storage depth the network module and the switch externally into
Row longitudinal direction capacity extension or lateral capacity extension;The longitudinal direction capacity extension are as follows: multiple super fusions one in same super aggregators
The data backup carried out between body machine and/or Data Migration;The transverse direction capacity extension are as follows: in the quantity of the super aggregators
When greater than one, the Data Migration that is carried out between different super aggregators.Super fusion cloud computing in the embodiment of the present invention
Computing module, network module, memory module and control module are arranged in the same super fusion all-in-one machine framework of system, i.e.,
On the same physical node, framework is simple, reduces hardware cost, while can calculate according in current super aggregators
The resource occupation information of module and memory module neatly carries out extending transversely or Longitudinal Extension, solves memory module performance
The problem of bottleneck, has stronger practicability and ease for use.
Embodiment two:
Fig. 5 shows the schematic diagram of super fusion cloud computing system provided by Embodiment 2 of the present invention.With reference to Fig. 5, this, which surpasses, melts
Closing cloud computing system includes switch and a super aggregators, and the super aggregators include three super fusion one
Machine;Described three super fusion all-in-one machines are communicatively coupled by the switch;Each super fusion all-in-one machine
Including computing module, network module, memory module and control module;The computing module is used for the computing module and storage
Resource service condition in module is analyzed, and corresponding resource occupation information is obtained;The resource service condition includes calculating
The operation utilization rate of module and the storage occupancy of memory module;The control module is used for true according to the resource occupation information
Whether the fixed super fusion all-in-one machine needs to carry out expanding storage depth, and passes sequentially through when needing to carry out expanding storage depth
The network module and the switch externally carry out longitudinal capacity extension or lateral capacity extension.
Wherein, longitudinal capacity extension are as follows: the number carried out between multiple super fusion all-in-one machines in same super aggregators
According to backup and/or Data Migration;The transverse direction capacity extension are as follows: when the quantity of the super aggregators is greater than one, not
With the Data Migration carried out between super aggregators.
Optionally, the computing module is for dividing the resource service condition in the computing module and memory module
Analysis, obtains corresponding resource occupation information, specifically includes:
Computing module is for determining resource comprehensive to the operation utilization rate of computing module and the storage occupancy of memory module
Occupancy, using resource comprehensive occupancy as corresponding resource occupation information.
It further, can be according to the needs of analysis, respectively the storage occupancy setting of operation utilization rate and memory module
Corresponding weight.For example, the weight of operation utilization rate is W1, the storage occupancy W2 of memory module, then resource comprehensive occupancy
=operation utilization rate * W1+ stores occupancy * W2.For example, operation utilization rate is 60% (i.e. 0.6), the weight of operation utilization rate is
W1=0.6, storage occupancy are 70% (i.e. 0.7), store occupancy W2=0.4, then resource comprehensive occupancy=0.6*0.6+
0.7*0.4=0.64.
Optionally, the control module be used for according to the resource occupation information determine the super fusion all-in-one machine whether need
Expanding storage depth is carried out, and passes sequentially through the network module and data friendship when needing to carry out expanding storage depth
Exchange device externally carries out longitudinal capacity extension or lateral capacity extension, specifically:
It include the first surpassing fusion all-in-one machine the second to surpass fusion all-in-one machine at least one in the same super aggregators;Institute
It states and the first surpasses in fusion all-in-one machine, control module is true according to the operation utilization rate of computing module and the storage occupancy of memory module
Determine resource comprehensive occupancy;When the resource comprehensive occupancy is greater than predetermined occupancy, control module passes through network module, institute
State switch and the network module for the second surpassing fusion all-in-one machine with it is described the second surpass merge the control mould of all-in-one machine
Block is communicated, when determining that the memory module for the second surpassing fusion all-in-one machine meets Longitudinal Extension condition, described first
In super fusion all-in-one machine, control module control memory module the second is surpassed by network module and the switch to described
It merges all-in-one machine and carries out Longitudinal Extension.
When it is described the second surpass fusion all-in-one machine memory module it is impossible to meet when Longitudinal Extension demand, the first surpass described
It merges in all-in-one machine, control module control memory module passes sequentially through network module and the switch and melts to outside is super
It is extending transversely to close node progress.
For example, super fusion cloud computing system includes two super aggregators, each super aggregators include 3 super fusions one
Body machine, the memory module of the super fusion all-in-one machine in each super aggregators is it is impossible to meet when Longitudinal Extension demand, then
Determine that the configuration of the memory module of the super fusion all-in-one machine in current super aggregators is unable to satisfy Longitudinal Extension demand.If
The memory module of super fusion all-in-one machine in one of them super aggregators it is impossible to meet when Longitudinal Extension demand, then judges
When whether the memory module of the super fusion all-in-one machine in another super aggregators can satisfy Longitudinal Extension demand, if another
It is impossible to meet Longitudinal Extension demands for the memory module of super fusion all-in-one machine in a super aggregators, then one super
Aggregators the first surpass in fusion all-in-one machine, and control module control memory module passes sequentially through network module and the data are handed over
Exchange device carries out extending transversely to external super aggregators.It is understood that if super fusion in another super aggregators
The memory module of all-in-one machine can satisfy Longitudinal Extension demand, then carries out Longitudinal Extension.
Wherein, longitudinal capacity extension specifically:
By the Data Migration in the memory module for the first surpassing fusion all-in-one machine to the memory module for the second surpassing fusion all-in-one machine
It is backed up, after the memory module capacity amplification for the first surpassing fusion all-in-one machine, then will backup to and described the second surpass fusion
The Data Migration of the memory module of all-in-one machine to it is described the first surpass fusion all-in-one machine in capacity amplification after memory module;It is same
It include described the first surpassing fusion all-in-one machine and at least one described the second surpasses fusion all-in-one machine in super aggregators.
Optionally, the migration is online thermophoresis.Wherein, the amplification of capacity includes but is not limited to CPU core number, memory
2T is inside saved as, the CPU core number after upgrading for example, the CPU core number of current super fusion all-in-one machine is 8 cores with the HardwareUpgring of disk
For 16 cores, 8T is inside saved as.
Wherein, the lateral capacity extension specifically:
By the Data Migration in the memory module for the first surpassing fusion all-in-one machine to the memory module for the second surpassing fusion all-in-one machine;
It is described the first surpass fusion all-in-one machine and it is described the second surpass fusion all-in-one machine do not belong to a super aggregators.
For example, super fusion cloud computing system includes a super aggregators A, super aggregators A includes 3 super fusion one
Machine a1, a2 and a3 need to increase at least one super aggregators according to extending transversely when needing extending transversely, it is assumed that increase
Adding super aggregators B, super aggregators B includes super fusion all-in-one machine b1, b2 and b3, then will merge all-in-one machine a1, a2 and/or a3
Memory module in Data Migration to it is super fusion all-in-one machine b1, b2 and/or b3 in memory module.It is understood that working as
It, can depositing one fusion all-in-one machine (a1, a2 or a3) in super aggregators A when demand extending transversely is the first demand
Store up module in Data Migration to it is super fusion all-in-one machine b1, b2 or b3 in memory module;When demand extending transversely is second
It, can be by two fusion all-in-one machines (a1, a2 in super aggregators A when demand;A1, a2 or a2, a3) memory module in
Data move to the memory module in super fusion all-in-one machine b1, b2 or b3 together;When demand extending transversely is third demand,
Data in the memory module of super fusion all-in-one machine (a1, a2 and a3) of three in super aggregators A can be moved to together
Memory module in super fusion all-in-one machine b1, b2 or b3;When demand extending transversely is four demand, super fusion can be saved
Two super fusion all-in-one machine (a1, a2 in point A;A1, a3 or a2, a3) memory module in data respectively correspond and move to two
A super fusion all-in-one machine (b1, b2;B1, b3 or b2, b3) in memory module, for example, by the super fusion one in super aggregators A
Data Migration in the memory module of body machine a1 melts super in super aggregators A to the memory module of super fusion all-in-one machine b1
Close memory module of the Data Migration in the memory module of all-in-one machine a2 to super fusion all-in-one machine b2, surpassing in super aggregators A
The data merged in the memory module of all-in-one machine a3 do not migrate.When demand extending transversely is five demand, can will surpass
Data in the memory module of three super fusions all-in-one machine (a1, a2 and a3) in aggregators A, which respectively correspond, moves to three
Memory module in super fusion all-in-one machine (b1, b2 and b3), for example, by the data point in the memory module of super fusion all-in-one machine a1
The memory module in super fusion all-in-one machine b1 Dui Ying not be moved to, the data in the memory module of super fusion all-in-one machine a2 are distinguished
The corresponding memory module moved in super fusion all-in-one machine b2 is right respectively by the data in the memory module of super fusion all-in-one machine a3
The memory module in super fusion all-in-one machine b3 should be moved to.
It should be noted that above-mentioned first demand, the second demand, third demand, the 4th demand and the 5th demand are different
Demand, optionally, the first demand, the second demand, third demand, the 4th demand and the 5th demand are sequentially increased.Optionally, institute
State the access frequency that computing module is also used to calculate data in the memory module.
The data that the control module is also used to by access frequency in the memory module lower than predeterminated frequency pass sequentially through
The network module and the switch are migrated to the memory module of external super fusion all-in-one machine.
Wherein, access frequency refers to the number accessed in preset a period of time.For example, predeterminated frequency is to access in 60 days
2 times, then access frequency (for example, accessing 1 time in 60 days) is lower than predeterminated frequency (accessing 2 times in 60 days) in the memory module
Data need to migrate, and are migrated by the network module and the switch to the storage of external super fusion all-in-one machine
Module.It should be noted that the memory module of external super fusion all-in-one machine refer to relative to current super fusion all-in-one machine it
Outer super fusion all-in-one machine, super fusion all-in-one machine including the same super aggregators and does not belong to the same super aggregators
Super fusion all-in-one machine.
Optionally, the memory module is multiple disks for storing data.
Optionally, the super aggregators include at least three surpassing fusion all-in-one machine.It is understood that each super fusion
Node can also include four super fusion all-in-one machines, or the super aggregators in part include three super fusion all-in-one machines, and part is super
Aggregators include four super fusion all-in-one machines, and this is not restricted.
The super fusion cloud computing system provided in the embodiment of the present invention, including switch and a super fusion section
Point, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion all-in-one machine is handed over by the data
Exchange device is communicatively coupled;Each super fusion all-in-one machine includes computing module, network module, memory module and control mould
Block;The computing module obtains corresponding for analyzing the resource service condition in the computing module and memory module
Resource occupation information;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module occupies
Rate;The control module is used to determine whether the super fusion all-in-one machine needs to carry out storage appearance according to the resource occupation information
Amount extension, and pass sequentially through when needing to carry out expanding storage depth the network module and the switch externally into
Row longitudinal direction capacity extension or lateral capacity extension;The longitudinal direction capacity extension are as follows: multiple super fusions one in same super aggregators
The data backup carried out between body machine and/or Data Migration;The transverse direction capacity extension are as follows: in the quantity of the super aggregators
When greater than one, the Data Migration that is carried out between different super aggregators.Super fusion cloud computing in the embodiment of the present invention
Computing module, network module, memory module and control module are arranged in the same super fusion all-in-one machine framework of system, i.e.,
On the same physical node, framework is simple, reduces hardware cost, while can calculate according in current super aggregators
The resource occupation information of module and memory module neatly carries out extending transversely or Longitudinal Extension, solves memory module performance
The problem of bottleneck, has stronger practicability and ease for use.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (7)
1. the super fusion cloud computing system of one kind, including switch, which is characterized in that further include at least one super fusion section
Point, each super aggregators include multiple super fusion all-in-one machines;The multiple super fusion all-in-one machine is handed over by the data
Exchange device is communicatively coupled;
Each super fusion all-in-one machine includes computing module, network module, memory module and control module;
The computing module obtains corresponding for analyzing the resource service condition in the computing module and memory module
Resource occupation information;The resource service condition includes that the storage of the operation utilization rate and memory module of computing module occupies
Rate;
The control module is used to determine whether the super fusion all-in-one machine is stored according to the resource occupation information
Capacity extension, and the network module is passed sequentially through when needing to carry out expanding storage depth and the switch is external
Carry out longitudinal capacity extension or lateral capacity extension;
The longitudinal direction capacity extension are as follows: in same super aggregators between multiple super fusion all-in-one machines the data backup that carries out and/
Or Data Migration;
The transverse direction capacity extension are as follows: when the quantity of the super aggregators is greater than one, between different super aggregators
The Data Migration carried out.
2. super fusion cloud computing system as described in claim 1, which is characterized in that the control module is used for according to the money
Source occupied information determines whether the super fusion all-in-one machine needs to carry out expanding storage depth, and is needing to carry out memory capacity expansion
The network module and the switch are passed sequentially through when exhibition externally carries out longitudinal capacity extension or lateral capacity extension,
Specifically:
It include the first surpassing fusion all-in-one machine the second to surpass fusion all-in-one machine at least one in the same super aggregators;Described
One surpass in fusion all-in-one machine, control module determines money according to the operation utilization rate of computing module and the storage occupancy of memory module
Source integrates occupancy;When the resource comprehensive occupancy is greater than predetermined occupancy, control module passes through network module, the number
According to switching equipment and the network module for the second surpassing fusion all-in-one machine with it is described the second surpass merge the control module of all-in-one machine into
Row communication, determine it is described the second surpass fusion all-in-one machine memory module meet Longitudinal Extension condition when, it is described the first surpass melt
It closes in all-in-one machine, control module control memory module the second surpasses fusion to described by network module and the switch
All-in-one machine carries out Longitudinal Extension;
When it is described the second surpass fusion all-in-one machine memory module it is impossible to meet when Longitudinal Extension demand, the first surpass fusion described
In all-in-one machine, control module control memory module passes sequentially through network module and the switch to the super fusion section in outside
Point carries out extending transversely.
3. super fusion cloud computing system as claimed in claim 1 or 2, which is characterized in that the longitudinal direction capacity extension specifically:
Data Migration in the memory module for the first surpassing fusion all-in-one machine is carried out to the memory module for the second surpassing fusion all-in-one machine
Backup, it is described the first surpass fusion all-in-one machine memory module capacity amplification after, then will backup to it is described the second surpass fusion one
The Data Migration of the memory module of machine to it is described the first surpass fusion all-in-one machine in capacity amplification after memory module;It is same super to melt
Close in node includes described the first surpassing fusion all-in-one machine and at least one described the second surpasses fusion all-in-one machine.
4. super fusion cloud computing system as claimed in claim 1 or 2, which is characterized in that the transverse direction capacity extension specifically:
By the Data Migration in the memory module for the first surpassing fusion all-in-one machine to the memory module for the second surpassing fusion all-in-one machine;It is described
The first surpass fusion all-in-one machine and the fusion all-in-one machine that the second surpasses does not belong to a super aggregators.
5. super fusion cloud computing system as described in claim 1, which is characterized in that the computing module is also used to calculate described
The access frequency of data in memory module;
The control module is also used to pass sequentially through access frequency in the memory module lower than the data of predeterminated frequency described
Network module and the switch are migrated to the memory module of external super fusion all-in-one machine.
6. super fusion cloud computing system as described in claim 1, which is characterized in that the memory module is multiple for storing
The disk of data.
7. super fusion cloud computing system as described in claim 1, which is characterized in that the super aggregators include at least three
Super fusion all-in-one machine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910145374.5A CN109951531B (en) | 2019-02-27 | 2019-02-27 | Super-fusion cloud computing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910145374.5A CN109951531B (en) | 2019-02-27 | 2019-02-27 | Super-fusion cloud computing system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109951531A true CN109951531A (en) | 2019-06-28 |
CN109951531B CN109951531B (en) | 2021-05-07 |
Family
ID=67008087
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910145374.5A Active CN109951531B (en) | 2019-02-27 | 2019-02-27 | Super-fusion cloud computing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109951531B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111488218A (en) * | 2020-03-13 | 2020-08-04 | 合肥中科类脑智能技术有限公司 | All-in-one machine computing system and method |
CN114143274A (en) * | 2021-11-16 | 2022-03-04 | 酒泉钢铁(集团)有限责任公司 | Method for integrating enterprise hyper-integration architecture in traditional architecture cloud platform |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105892952A (en) * | 2016-04-22 | 2016-08-24 | 深圳市深信服电子科技有限公司 | Hyper-converged system and longitudinal extension method thereof |
CN107037985A (en) * | 2017-02-14 | 2017-08-11 | 中山大学 | The super fusion integrated machine system of one kind and its horizontal and vertical expansion method |
CN107132990A (en) * | 2016-02-26 | 2017-09-05 | 深圳市深信服电子科技有限公司 | Reading I O scheduling method and device based on super fusion storage |
CN107426012A (en) * | 2017-05-27 | 2017-12-01 | 深信服科技股份有限公司 | A kind of fault recovery method and its device based on super fusion architecture |
US20190012092A1 (en) * | 2017-07-05 | 2019-01-10 | Lenovo Enterprise Solutions (Singapore) Pte. Ltd. | Managing composable compute systems with support for hyperconverged software defined storage |
-
2019
- 2019-02-27 CN CN201910145374.5A patent/CN109951531B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107132990A (en) * | 2016-02-26 | 2017-09-05 | 深圳市深信服电子科技有限公司 | Reading I O scheduling method and device based on super fusion storage |
CN105892952A (en) * | 2016-04-22 | 2016-08-24 | 深圳市深信服电子科技有限公司 | Hyper-converged system and longitudinal extension method thereof |
CN107037985A (en) * | 2017-02-14 | 2017-08-11 | 中山大学 | The super fusion integrated machine system of one kind and its horizontal and vertical expansion method |
CN107426012A (en) * | 2017-05-27 | 2017-12-01 | 深信服科技股份有限公司 | A kind of fault recovery method and its device based on super fusion architecture |
US20190012092A1 (en) * | 2017-07-05 | 2019-01-10 | Lenovo Enterprise Solutions (Singapore) Pte. Ltd. | Managing composable compute systems with support for hyperconverged software defined storage |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111488218A (en) * | 2020-03-13 | 2020-08-04 | 合肥中科类脑智能技术有限公司 | All-in-one machine computing system and method |
CN114143274A (en) * | 2021-11-16 | 2022-03-04 | 酒泉钢铁(集团)有限责任公司 | Method for integrating enterprise hyper-integration architecture in traditional architecture cloud platform |
Also Published As
Publication number | Publication date |
---|---|
CN109951531B (en) | 2021-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103503414B (en) | A kind of group system calculating storage and merge | |
CN105978704B (en) | System and method for creating new cloud resource instruction set architecture | |
CN105468473B (en) | Data migration method and data migration device | |
CN103631538B (en) | Cold and hot data identification threshold value calculation, device and system | |
CN100367727C (en) | Expandable storage system and control method based on objects | |
CN103763383B (en) | Integrated cloud storage system and its storage method | |
CN104113597B (en) | The HDFS data read-write method of a kind of many Data centres | |
CN103034664B (en) | Control database the method, system and device of Data Migration | |
CN102255962A (en) | Distributive storage method, device and system | |
CN102982182B (en) | Data storage planning method and device | |
CN107465656A (en) | A kind of safety monitoring big data processing method and system based on cloud computing | |
CN104461685B (en) | Virtual machine processing method and virtual computer system | |
CN104735110B (en) | Metadata management method and system | |
CN105808443B (en) | A kind of method, apparatus and system of Data Migration | |
CN103778212B (en) | Parallel mass data processing method based on back end | |
CN109951531A (en) | Super fusion cloud computing system | |
CN205490680U (en) | High available cloud computing system based on general server and universal exchange | |
CN106919346A (en) | A kind of shared Storage Virtualization implementation method based on CLVM | |
CN108874590A (en) | A kind of system of cloud host automated back-up and recovery | |
CN110489407A (en) | Data filling mining method, apparatus, computer equipment and storage medium | |
CN102844772A (en) | Managing committed request rates for shared resources | |
CN108052569A (en) | Data bank access method, device, computer readable storage medium and computing device | |
CN107943602A (en) | Hardware abstraction plateform system and equipment based on AVS2 codings | |
CN105373452B (en) | A kind of data back up method | |
CN106227464A (en) | A kind of double-deck redundant storage system and data write, reading and restoration methods |
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 |