CN109033007A - A kind of dynamic computing device - Google Patents
A kind of dynamic computing device Download PDFInfo
- Publication number
- CN109033007A CN109033007A CN201810664354.4A CN201810664354A CN109033007A CN 109033007 A CN109033007 A CN 109033007A CN 201810664354 A CN201810664354 A CN 201810664354A CN 109033007 A CN109033007 A CN 109033007A
- Authority
- CN
- China
- Prior art keywords
- circuit
- computing device
- unit
- reconfigurable circuit
- dynamic computing
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7867—Architectures of general purpose stored program computers comprising a single central processing unit with reconfigurable architecture
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Abstract
The present invention discloses a kind of Dynamic calculation method, is related to technical field of data processing, for improving the data-handling capacity of computer.The Dynamic calculation method includes: computing capability required for estimating cloud storage；According to the required computing capability estimated out, reconfigurable circuit is constructed；The reconfigurable circuit of building is formed into virtual processor；Virtual processor is distributed into corresponding virtual machine；Calculating needed for completing cloud storage as virtual machine.The present invention is suitable for computer field.
Description
The application is application No. is 2017101101003, and the applying date is on February 28th, 2017, and invention and created name is
A kind of divisional application of the patent of " Dynamic calculation method ".
Technical field
The present invention relates to technical field of data processing more particularly to a kind of Dynamic calculation methods.
Background technique
With the development of science and technology, all trades and professions, which require computer, a series of operation such as is stored, is calculated, this requirement
Computer has powerful computing capability and processing processing capacity.Especially, cloud storage is as new fashion trend, it is often necessary to
Carry out a large amount of data storage and calculating.
In the prior art, by increasing the quantity of processor, increase processing capacity required for cloud storage, although calculating effect
Rate increases, but calculates increased costs.
Summary of the invention
The purpose of the present invention is to provide a kind of Dynamic calculation method and its devices, for not increasing the same of calculating cost
When, improve the data-handling capacity of computer.
In order to achieve the above objectives, the present invention adopts the following technical scheme:
The first aspect of the present invention provides a kind of Dynamic calculation method, which is characterized in that the Dynamic calculation method includes:
Step 1: estimating computing capability required for cloud storage.
Step 2: constructing reconfigurable circuit according to the required computing capability estimated out.
Step 3: the reconfigurable circuit of building is formed virtual processor.
Step 4: virtual processor is distributed to corresponding virtual machine.
Step 5: calculating needed for completing cloud storage as virtual machine.
According to above-mentioned Dynamic calculation method, computing capability required for cloud storage is estimated first, according to computer
Group and/or the existing computing capability of computer, are allocated processor from newly, thus achieve the purpose that quickly to handle data,
Constantly increase processor in compared with the prior art, this programme not only calculates cost without increasing, and effectively mentions
High computational efficiency.
The Dynamic calculation method further include: between step 1 and step 2, count the quantity of idle logic circuit.
In step 2, reconfigurable circuit includes n idle logic circuit, wherein n is positive integer；If n is equal to 1, select
Select any one logic circuit composition reconfigurable circuit being in idle condition；If n is greater than 1, select physical address adjacent
Two or more logic circuit composition reconfigurable circuits being in idle condition.
N logic circuit for forming reconfigurable circuit is arranged to possession state.
Logic circuit is coprocessor circuit, fast Fourier transform circuit, surmounts function circuit, bus control unit electricity
Road, timer circuit or hardware algorithm build modular circuit.
In step 4, virtual processor is one-to-one relationship with virtual machine；Alternatively, multiple virtual processors corresponding one
A virtual machine.
Step 5 further includes, if mistake occurs in virtual machine calculating process, terminates the calculating process, and by the calculating process
The content of involved reconfigurable circuit is removed, and so that reconfigurable circuit is restored to idle state, and rebuild reconfigurable circuit.
The Dynamic calculation method further include: Step 6: after calculating terminates or terminate, the content in reconfigurable circuit is clear
It removes, reconfigurable circuit is made to be restored to idle state.
The second aspect of the present invention provides a kind of dynamic computing device, which includes: to estimate unit, is used for
Estimate computing capability required for cloud storage；With the construction unit estimating unit and being connected, by according to estimate out it is required based on
Calculation ability constructs reconfigurable circuit；The processing unit being connected with construction unit, the virtual processing of reconfigurable circuit composition for that will construct
Device；The allocation unit being connected with processing unit, for virtual processor to be distributed to corresponding virtual machine.
The dynamic computing device further include: with the statistic unit estimating unit and construction unit and being respectively connected with, for counting
N idle logic circuit is formed reconfigurable circuit, wherein n is positive integer by the quantity of logic circuit.
The dynamic computing device further include: clearing cell is used for after calculating terminates or terminate, will be in reconfigurable circuit
Content is removed, and reconfigurable circuit is made to be restored to idle state.
According to above-mentioned dynamic computing device, processor is allocated again, thus achieve the purpose that quickly to handle data,
Constantly increase the solution of processor in compared with the prior art, this programme not only calculates cost, Er Qieyou without increasing
Improve computational efficiency to effect.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, embodiment will be described below
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 creative efforts, can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is the flow chart of Dynamic calculation method in embodiment；
Fig. 2 is the structural schematic diagram of dynamic computing device in embodiment.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Embodiment one
A kind of Dynamic calculation method is present embodiments provided, as shown in Figure 1, the Dynamic calculation method includes:
Step 1: estimating computing capability required for cloud storage.
Step 2: constructing reconfigurable circuit according to the required computing capability estimated out.
Step 3: the reconfigurable circuit of building is formed virtual processor.
Step 4: virtual processor is distributed to corresponding virtual machine.
Step 5: calculating needed for completing cloud storage as virtual machine.
According to above-mentioned Dynamic calculation method, computing capability required for cloud storage is estimated first, according to computer
Group and/or the existing computing capability of computer, are again allocated processor, thus achieve the purpose that quickly to handle data,
Constantly increase the solution of processor in compared with the prior art, this programme not only calculates cost, Er Qieyou without increasing
Improve computational efficiency to effect.
In order to clearly illustrate Dynamic calculation method in the present embodiment, above-mentioned Dynamic calculation method is carried out below detailed
Description:
Firstly, according to the main control unit in computer cluster and/or computer to needing meter required for carrying out cloud storage
Calculation ability is estimated, and certainly, which includes all processors and port corresponding with all processors.
It is preferred in this example, by main control unit, any one idle processor or multiple idle processor compositions it is whole
Body needs computing capability required for carrying out cloud storage to estimate to above-mentioned.
Then, according to the above-mentioned computing capability estimated, reconfigurable circuit is constructed, which can be logic circuit, show
Example property, it can be coprocessor circuit, fast Fourier transform circuit, surmount function circuit, bus control unit circuit, timer
Circuit or hardware algorithm build the combination of one or more of modular circuit.
Illustratively, the present embodiment can also count the quantity of idle logic circuit, and will before constructing reconfigurable circuit
Idle logic circuit forms reconfigurable circuit.Specifically, the Dynamic calculation method further include: between step 1 and step 2,
Count the quantity of idle logic circuit, the computing capability according to needed for cloud storage, the electricity of logic needed for counting the computing capability
Road, and that idle logic circuit is formed into reconfigurable circuit, wherein n is positive integer.
Also, according to required computing capability, the quantity of n can be 1 or multiple, specifically, if selection is appointed when n is equal to 1
The logic circuit composition reconfigurable circuit that meaning one is in idle condition；If n is greater than 1, two that physical address is adjacent are selected
Or multiple logic circuit composition reconfigurable circuits being in idle condition.
In order to not be disturbed in data processing, n logic circuit for forming reconfigurable circuit can be arranged to occupy
State.Wherein, which can be coprocessor circuit, fast Fourier transform circuit, surmount function circuit, bus
Controller circuitry, timer circuit or hardware algorithm build modular circuit.
Further, the reconfigurable circuit built is formed into virtual processor, and virtual processor is distributed to relatively
The virtual machine answered.Calculating needed for completing cloud storage as virtual machine.At this point, the corresponding relationship of virtual processor and virtual machine can be
One-to-one relationship；Alternatively, the corresponding relationship of virtual processor and virtual machine can be many-to-one relationship.Above-mentioned corresponding relationship
Selection, mainly depend on the quantity with required computing capability and the virtual machine that can be called.For example, if in the calculating process
In, only one virtual machine can call, then the corresponding relationship of above-mentioned virtual processor and virtual machine is then to a virtual processor
The corresponding same virtual machine.For another example in the calculating process, the virtual machine quantity that can be called be it is multiple, will can virtually handle
The corresponding relationship of device and virtual machine is to correspond, and is conducive to the processing speed for improving data in this way, to effectively improve number
According to processing capacity.
Certainly, in step 5, if mistake occurs in calculating process in virtual machine, the calculating process is terminated, and should
The content of reconfigurable circuit involved by calculating process is removed, and so that reconfigurable circuit is restored to idle state, and rebuild reconstruct
Circuit.That is, if occurring mistake in calculating process and being returned in step 2, rebuild reconfigurable circuit.To avoid
Malfunction due to calculating, caused by calculate overlong time, further increase computational efficiency, save and calculate time.
Alternatively, watchdog circuit can also be arranged on a virtual machine, if virtual machine does not export feeding-dog signal for a long time, terminate
Calculating process, and the content for the reconfigurable circuit being involved in is removed, and reconfigurable circuit is made to be restored to idle state, and rebuilds weight
Structure circuit.
Illustratively, Dynamic calculation method in this implementation further include: Step 6: will be weighed after calculating terminates or terminate
Content in structure circuit is removed, and reconfigurable circuit is made to be restored to idle state.To which the reconfigurable circuit occupied in calculating process be released
It puts, convenient for completing calculating next time.
Inevitably, in the present embodiment data processing complete after, be just stored in server or relative to memory in,
Illustratively, which can be flash storage, i.e., novel nonvolatile memory can be effectively prevented and count after a loss of power
According to loss, the access speed of data is fast.
Embodiment two
The present embodiment provides a kind of dynamic computing device, which is suitable for moving described in above-described embodiment one
State calculation method, as shown in Fig. 2, the dynamic computing device includes:
Unit 1 is estimated, for estimating computing capability required for cloud storage；With the construction unit 2 estimating unit 1 and being connected,
For constructing reconfigurable circuit according to calculated required computing capability；The processing unit 3 being connected with construction unit 2, being used for will
The reconfigurable circuit of building forms virtual processor；The allocation unit 4 being connected with processing unit 3, for distributing virtual processor
To corresponding virtual machine.
The dynamic computing device may also include that and estimate the statistic unit that unit 1 and construction unit 2 are respectively connected with, and be used for
N idle logic circuit is formed reconfigurable circuit, wherein n is positive integer by the quantity of statistic logic circuit.
Further, which connect with one or more virtual machines, these virtual machines are deposited for completing cloud
Calculating needed for storage, dynamic computing device are that these virtual machines distribute virtual processor.
Above-mentioned virtual machine can be computer cluster and/or computer.
In addition, the dynamic computing device may also include that watchdog unit, which is made of watchdog chip,
The watchdog unit receives the regular and/or irregular feeding-dog signal that virtual processor is sent back to, if watchdog unit does not receive
Regular and/or irregular feeding-dog signal, then terminate calculating process；And the termination signal is sent to clearing cell, it removes single
After member receives the signal, the content for the reconfigurable circuit being related in calculating process is removed, reconfigurable circuit is made to be restored to idle shape
State.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (10)
1. a kind of dynamic computing device, which is characterized in that the dynamic computing device includes:
Unit is estimated, for estimating computing capability required for cloud storage；
Construction unit, for constructing reconfigurable circuit according to calculated required computing capability；
Processing unit, the reconfigurable circuit for that will construct form virtual processor；
Allocation unit, for virtual processor to be distributed to corresponding virtual machine；
Statistic unit, the quantity for statistic logic circuit.
2. dynamic computing device according to claim 1, which is characterized in that the reconfigurable circuit includes n idle logic
Circuit, wherein n is positive integer；
If n is equal to 1, which is made of the logic circuit that any one is in idle condition；If n is greater than 1,
The reconfigurable circuit is made of the adjacent two or more logic circuits being in idle condition of physical address.
3. dynamic computing device according to claim 2, which is characterized in that the statistic unit and estimate unit and building singly
Member is respectively connected with.
4. dynamic computing device according to claim 3, which is characterized in that the statistic unit is electric by n idle logic
Road forms reconfigurable circuit, wherein n is positive integer.
5. dynamic computing device according to claim 4, which is characterized in that the dynamic computing device further include: setting is single
Member, the setting unit are used to n logic circuit for forming reconfigurable circuit being arranged to possession state.
6. Dynamic calculation method according to claim 5, which is characterized in that the logic circuit is coprocessor circuit, fastly
Fast Fourier-transform circuitry surmounts function circuit, bus control unit circuit, timer circuit and hardware algorithm construction modular circuit
One or more of combinations.
7. dynamic computing device according to claim 1, which is characterized in that the dynamic computing device further include: remove single
Member, for the content in the reconfigurable circuit being removed, the reconfigurable circuit is made to be restored to idle shape after calculating terminating or terminate
State.
8. dynamic computing device according to claim 7, which is characterized in that the dynamic computing device, which may also include that, guards the gate
Dog unit, the watchdog unit are made of watchdog chip, which is used to receive that the virtual processor to send back to determines
Phase and/or irregular feeding-dog signal.
9. dynamic computing device according to claim 8, which is characterized in that if the watchdog unit do not receive periodically and/
Or irregular feeding-dog signal, then termination signal is issued, calculating process is terminated.
10. dynamic computing device according to claim 9, which is characterized in that the watchdog unit is also by the termination signal
It is sent to the clearing cell, it is after which receives the signal, the content for the reconfigurable circuit being related in calculating process is clear
It removes, the reconfigurable circuit is made to be restored to idle state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810664354.4A CN109033007A (en) | 2017-02-28 | 2017-02-28 | A kind of dynamic computing device |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810664354.4A CN109033007A (en) | 2017-02-28 | 2017-02-28 | A kind of dynamic computing device |
| CN201710110100.3A CN106909530B (en) | 2017-02-28 | 2017-02-28 | A kind of Dynamic calculation method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710110100.3A Division CN106909530B (en) | 2017-02-28 | 2017-02-28 | A kind of Dynamic calculation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109033007A true CN109033007A (en) | 2018-12-18 |
Family
ID=59208062
Family Applications (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710110100.3A Expired - Fee Related CN106909530B (en) | 2017-02-28 | 2017-02-28 | A kind of Dynamic calculation method |
| CN201810664354.4A Withdrawn CN109033007A (en) | 2017-02-28 | 2017-02-28 | A kind of dynamic computing device |
| CN201810201331.XA Active CN108415876B (en) | 2017-02-28 | 2017-02-28 | Dynamic calculation device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710110100.3A Expired - Fee Related CN106909530B (en) | 2017-02-28 | 2017-02-28 | A kind of Dynamic calculation method |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810201331.XA Active CN108415876B (en) | 2017-02-28 | 2017-02-28 | Dynamic calculation device |
Country Status (1)
| Country | Link |
|---|---|
| CN (3) | CN106909530B (en) |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9137304B2 (en) * | 2011-05-25 | 2015-09-15 | Alcatel Lucent | Method and apparatus for achieving data security in a distributed cloud computing environment |
| CN102624870A (en) * | 2012-02-01 | 2012-08-01 | 北京航空航天大学 | Intelligent optimization algorithm based cloud manufacturing computing resource reconfigurable collocation method |
| CN102722413B (en) * | 2012-05-16 | 2017-12-29 | 上海兆民云计算科技有限公司 | The distributed resource scheduling method that a kind of desktop cloud cluster uses |
| US9251115B2 (en) * | 2013-03-07 | 2016-02-02 | Citrix Systems, Inc. | Dynamic configuration in cloud computing environments |
| WO2014190094A1 (en) * | 2013-05-21 | 2014-11-27 | Ecrio, Inc. | Real-time rich communications client architecture |
| US9317313B2 (en) * | 2013-05-22 | 2016-04-19 | Microsoft Technology Licensing, Llc | Dynamically provisioning storage while identifying and discarding redundant storage alerts |
| CN103365727B (en) * | 2013-07-09 | 2017-02-15 | 南京大学 | Host load forecasting method in cloud computing environment |
| CN103455378A (en) * | 2013-09-05 | 2013-12-18 | 华为技术有限公司 | Resource distribution method of virtual clusters, device and system |
| WO2015058377A1 (en) * | 2013-10-23 | 2015-04-30 | 华为技术有限公司 | Method and device for creating virtual machine |
| CN103678004A (en) * | 2013-12-19 | 2014-03-26 | 南京大学 | Host load prediction method based on unsupervised feature learning |
| CN104915151B (en) * | 2015-06-02 | 2018-12-07 | 杭州电子科技大学 | A kind of memory excess distribution method that active is shared in multi-dummy machine system |
| CN105007311A (en) * | 2015-07-03 | 2015-10-28 | 叶秀兰 | System and method for resource management based on cloud platform and cloud computing |
| CN106326002B (en) * | 2015-07-10 | 2020-10-20 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device and equipment |
| CN106250217A (en) * | 2016-07-22 | 2016-12-21 | 无锡华云数据技术服务有限公司 | Synchronous dispatching method between a kind of many virtual processors and dispatching patcher thereof |
| CN106407015A (en) * | 2016-10-13 | 2017-02-15 | 安徽鼎集信息科技有限公司 | High-reliability virtualization server system |
| CN106357800B (en) * | 2016-10-21 | 2020-06-05 | 黄东 | QoE-based cloud computing service architecture |
-
2017
- 2017-02-28 CN CN201710110100.3A patent/CN106909530B/en not_active Expired - Fee Related
- 2017-02-28 CN CN201810664354.4A patent/CN109033007A/en not_active Withdrawn
- 2017-02-28 CN CN201810201331.XA patent/CN108415876B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108415876A (en) | 2018-08-17 |
| CN106909530A (en) | 2017-06-30 |
| CN106909530B (en) | 2018-06-05 |
| CN108415876B (en) | 2020-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102891896B (en) | Cloud computing infrastructure system for mass storage of Internet of things | |
| CN106951926A (en) | The deep learning systems approach and device of a kind of mixed architecture | |
| CN103455378A (en) | Resource distribution method of virtual clusters, device and system | |
| CN112003797B (en) | Method, system, terminal and storage medium for improving performance of virtualized DPDK network | |
| Dong et al. | Virtual machine placement for improving energy efficiency and network performance in iaas cloud | |
| CN103902459A (en) | Method and associated equipment for determining management mode of shared virtual memory page | |
| CN103336722A (en) | Virtual machine CPU source monitoring and dynamic distributing method | |
| CN108519917A (en) | A kind of resource pool distribution method and device | |
| CN103279330A (en) | MapReduce multiple programming model based on virtual machine GPU computation | |
| CN102136993A (en) | Data transfer method, device and system | |
| CN102637138A (en) | Method for computing and scheduling virtual machine | |
| CN103956991B (en) | A kind of FIR filtering Parallel Realization methods based on CPU/GPU heterogeneous platforms | |
| CN103309745A (en) | Method and device for distributing virtual resource in cloud architecture | |
| CN109587072A (en) | Distributed system overall situation speed limiting system and method | |
| WO2020108536A1 (en) | Virtual network resource allocation method and system and electronic device | |
| CN106354354A (en) | Control method and control device for capacitive touch screen as well as terminal equipment | |
| CN106909530B (en) | A kind of Dynamic calculation method | |
| Subramoni et al. | Design of network topology aware scheduling services for large infiniband clusters | |
| CN103329059A (en) | Circuitry to select, at least in part, at least one memory | |
| CN110119375A (en) | A kind of control method that multiple scalar cores are linked as to monokaryon Vector Processing array | |
| CN111338579A (en) | Read-write cache optimization method, system, terminal and storage medium based on storage pool | |
| CN106296279A (en) | Advertisement placement method and device | |
| CN106506594B (en) | A kind of concurrent computation resource distribution method and device | |
| CN103226495B (en) | One can switch course allocation method and equipment | |
| CN109460535A (en) | A kind of finite field matrix inversion device based on cloud and inversion technique |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181218 |