CN107037987A - To the treating method and apparatus rolled up in distributed memory system - Google Patents

To the treating method and apparatus rolled up in distributed memory system Download PDF

Info

Publication number
CN107037987A
CN107037987A CN201710213932.8A CN201710213932A CN107037987A CN 107037987 A CN107037987 A CN 107037987A CN 201710213932 A CN201710213932 A CN 201710213932A CN 107037987 A CN107037987 A CN 107037987A
Authority
CN
China
Prior art keywords
node
data
client
memory system
distributed memory
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
Application number
CN201710213932.8A
Other languages
Chinese (zh)
Inventor
王劲凯
游峰
亓武强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Internet Science And Technology Ltd Of Cloud Of China
Original Assignee
Beijing Internet Science And Technology Ltd Of Cloud Of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Internet Science And Technology Ltd Of Cloud Of China filed Critical Beijing Internet Science And Technology Ltd Of Cloud Of China
Priority to CN201710213932.8A priority Critical patent/CN107037987A/en
Publication of CN107037987A publication Critical patent/CN107037987A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/061—Improving I/O performance
    • G06F3/0611—Improving I/O performance in relation to response time
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
    • G06F3/065—Replication mechanisms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0662—Virtualisation aspects
    • G06F3/0665—Virtualisation aspects at area level, e.g. provisioning of virtual or logical volumes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiments of the invention provide a kind for the treatment of method and apparatus to being rolled up in distributed memory system.Methods described includes：It is that client creates a volume, the physics space-consuming of the volume is the predefined size of the volume in distributed memory system；It is that client selects a first node as the corresponding localization node of the client in the node of distributed memory system；It is that client selects at least one Section Point as the corresponding strange land node of the client in the node of distributed memory system；It is the first address that the client distributes predefined size in the localization node；It is the address that the volume distributes the predefined size at least one described strange land node；The client is received to first write data requests permed and sent；A complete copy data of the described first data to be written are write in the first node；By the complete copy data of the predetermined quantity of the described first data to be written, distribution is write in the strange land node.

Description

To the treating method and apparatus rolled up in distributed memory system
Technical field
The present invention relates to distributed memory system technical field, more particularly to a kind of place to being rolled up in distributed memory system Manage method and apparatus.
Background technology
In traditional distributed storage based on HDD (Hard Disk Drive, hard disk drive), using uniformity Hash Hash deposits data, and the copy of book is scattered is stored in each distributed node.By Data Concurrent access come The IO performances of lifting system.However, in NVME (Non-Volatile Memory Express, nonvolatile memory) SSD magnetic Today of disk high speed development, the SSD hard disks of individual node can provide very high IO performances, and cross-node is accessed and added on the contrary The delay of data access, it is impossible to play SSD performance completely.
The content of the invention
The embodiment provides a kind for the treatment of method and apparatus to being rolled up in distributed memory system, it can accelerate Client reads the reading speed for the data rolled up in distributed memory system.
To achieve these goals, this invention takes following technical scheme.
A kind of processing method to being rolled up in distributed memory system, including：
Step 1, it is that client creates a volume, the physics space-consuming of the volume is institute in distributed memory system State the predefined size of volume；
Step 2, it is that client selects a first node as the client pair in the node of distributed memory system The localization node answered；It is that client selects at least one Section Point as described in the node of distributed memory system The corresponding strange land node of client；
Step 3, it is the first address that the client distributes predefined size, for storing in the localization node One complete copy data of the volume；
Step 4, it is the address that the volume distributes the predefined size, for dividing at least one described strange land node Cloth stores the complete copy data of the predetermined quantity of the volume；
Step 5, the client is received to first write data requests permed and sent, the first write data requests bag Include the first data to be written；
Step 6, by a complete copy data of the described first data to be written, write in the first node；
Step 7, by the complete copy data of the predetermined quantity of the described first data to be written, the distribution write-in strange land section Point in.
A kind of processing method to being rolled up in distributed memory system, including：
Step 1, it is that client creates a volume, the virtual space-consuming of the volume is institute in distributed memory system The predefined size of volume is stated, the initial physical space-consuming of the volume is 0；
Step 2, the client is received to first write data requests permed and sent, the first write data requests bag Include the first data to be written；
Step 3, it is that the client selects a first node as the client in the node of distributed memory system Corresponding localization node is held, the complete copy data for storing the described first data to be written；
Step 4, it is that the client selects at least one node as the corresponding strange land node of the client, for dividing Cloth stores the complete copy data of the predetermined quantity of first data to be written；
Step 5, it will be localized described in a complete copy data Cun Chudao of the described first data to be written in node；By institute The complete copy distributed data storage of predetermined quantity of the first data to be written is stated into the localization node.
A kind of processing unit to being rolled up in distributed memory system, including：
Creating unit, is that client creates a volume, the physics space-consuming of the volume is in distributed memory system The predefined size of the volume；
Selecting unit, is that client selects a first node as the client in the node of distributed memory system Hold corresponding localization node；It is that client selects at least one Section Point conduct in the node of distributed memory system The corresponding strange land node of the client；
First allocation unit, in the localization node, is the first address that the client distributes predefined size, uses In the complete copy data for storing the volume；
Second allocation unit, is the address that the volume distributes the predefined size at least one described strange land node, For the complete copy data for the predetermined quantity for being distributed the storage volume；
Receiving unit, receives the client to first write data requests permed and sent, and described first writes data please Ask including the first data to be written；
First writing unit, by a complete copy data of the described first data to be written, writes the first node In；
Second writing unit, by the complete copy data of the predetermined quantity of the described first data to be written, distribution write-in institute State in the node of strange land.
A kind of operation device to being rolled up in distributed memory system, including：
Creating unit, is that client creates a volume, the virtual space-consuming of the volume is in distributed memory system The predefined size of the volume, the initial physical space-consuming of the volume is 0；
Receiving unit, receives the client to first write data requests permed and sent, and described first writes data please Ask including the first data to be written；
First choice unit, is that the client selects a first node conduct in the node of distributed memory system The corresponding localization node of the client, the complete copy data for storing the described first data to be written；
Second selecting unit, is that the client selects at least one node to be saved as the corresponding strange land of the client Point, the complete copy data of the predetermined quantity for being distributed storage first data to be written；
Memory cell, will be localized in node described in a complete copy data Cun Chudao of the described first data to be written； By the complete copy distributed data storage of the predetermined quantity of the described first data to be written into the localization node.
It is that client exists it can be seen from the technical scheme provided by embodiments of the invention described above in the embodiment of the present invention A volume is created in distributed memory system, a complete copy data of the volume is set on one node, works as client When end accesses the volume of distributed memory system, the localization node for the complete copy data for preserving volume can be connected, so that plus Access speed of the fast client to volume.
The additional aspect of the present invention and advantage will be set forth in part in the description, and these will become from the following description Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, being used required in being described below to embodiment Accompanying drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this For the those of ordinary skill of field, without having to pay creative labor, other can also be obtained according to these accompanying drawings Accompanying drawing.
Fig. 1 for one embodiment of the invention provide it is a kind of to the handling process for the processing method rolled up in distributed memory system Figure；
Fig. 2 for another embodiment of the present invention provide it is a kind of to the processing stream for the processing method rolled up in distributed memory system Cheng Tu；
A kind of structure for processing unit to being rolled up in distributed memory system that Fig. 3 provides for another embodiment of the present invention connects Connect flow chart；
A kind of structure for processing unit to being rolled up in distributed memory system that Fig. 4 provides for another embodiment of the present invention connects Connect flow chart.
Embodiment
Embodiments of the present invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning Same or similar element or element with same or like function are represented to same or similar label eventually.Below by ginseng The embodiment for examining accompanying drawing description is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.
The big small mode that following tupe is rolled up for predistribution, that is to say, that the initial physical space-consuming of volume is volume Predefined size.
As shown in figure 1, be a kind of operating method to being rolled up in distributed memory system of the present invention, including：
Step 11, it is that client creates a volume, the physics space-consuming of the volume is institute in distributed memory system State the predefined size of volume；
Step 12, it is that client selects a first node as the client in the node of distributed memory system Corresponding localization node；
Step 12 is specially：The first node for the distributed memory system that the client is currently connected is obtained, is made For localization node；In the embodiment, the node that can be currently connected according to client is used as localization node.
Or, step 12 is specially：It is that the client specifies one first in the distributed memory system interior joint Node, is used as localization node.In the embodiment, a node is specified as localization node, subsequently can be by local side Node address is sent to client so that client is directly connected to localize node.The step is specially：According to the distribution The performance of the current appearance load rate of the node of storage system, each node in the distributed memory system, and/or connected visitor The quantity at family end, in the distributed memory system interior joint, is that the client specifies a first node, is used as localization Node.Localization node is specified according to above-mentioned strategy, it is contemplated that the factor of each side, is more beneficial for the stabilization of system and is Equilibrium between system.
Step 13, it is that client selects at least one Section Point as described in the node of distributed memory system The corresponding strange land node of client；
Step 13 is specially：According to current appearance load rate, the distributed storage of the node of the distributed memory system The performance of each node in system, is that client selects at least one Section Point, is used as the corresponding strange land of the client Node.According to the specified strange land node of above-mentioned strategy, it is contemplated that the factor of each side, is more beneficial for the stabilization and system of system Between equilibrium.
Step 14, it is the first address that the client distributes predefined size, for storing in the localization node One complete copy data of the volume；
Step 15, it is the address that the volume distributes the predefined size, for dividing at least one described strange land node Cloth stores the complete copy data of the predetermined quantity of the volume；
Step 16, the client is received to first write data requests permed and sent, first write data requests Including the first data to be written；
Step 17, by a complete copy data of the described first data to be written, write in the first node；
Step 18, by the complete copy data of the predetermined quantity of the described first data to be written, the distribution write-in strange land In node.
It is that client creates a volume in distributed memory system, one of the volume is complete in the embodiment of the present invention Copy data set on one node, when client access distributed memory system volume when, can connect and preserve volume Complete copy data localization node, so as to accelerate access speed of the client to volume.
Methods described also includes：
Step 19, it is each in the distributed memory system according to all copy datas of the described first data to be written The distribution of individual node, records the storage address letter of each node of the copy data of the volume in the distributed memory system Breath.
Methods described also includes：
Step 110, when receiving that the client sends to the operational orders of the described first data to be written, according to institute Operational order is stated, the copy data to the described first data to be written in the strange land node is grasped according to the operational order Make；
Step 111, when the copy data in the strange land node of the described first data to be written refers to according to the operation Make after the completion of being operated, according to the operational order to the described first data to be written in the number of copies for localizing node According to being operated.
Optionally, between the step 11 and step 12, methods described also includes：
In the localization node, volume controller is created；The volume controller is used for the pair for recording the localization volume The storage address information of node of the notebook data in the distributed memory system, and for controlling to read the copy data Data consistency and control the IO orders of the copy data；
Accordingly, the step 19 is specially：The volume controller is according to all copies of the described first data to be written The distribution of each node of the data in the distributed memory system, the copy data for recording the volume is deposited in the distribution The storage address information of each node in storage system.
Second of pattern is described below.It is that client creates a volume, the virtual space-consuming of the volume is in the pattern The predefined size of the volume, during the initial physical space-consuming of the volume is 0, subsequent treatment, with continuous deposit data in volume, The physics space-consuming of volume is continuously increased.
As shown in Fig. 2 be a kind of operating method to being rolled up in distributed memory system of the present invention, including：
Step 21, it is that client creates a volume, the virtual space-consuming of the volume is institute in distributed memory system The predefined size of volume is stated, the initial physical space-consuming of the volume is 0；
Step 22, the client is received to first write data requests permed and sent, first write data requests Including the first data to be written；
Step 23, it is that the client selects a first node as the visitor in the node of distributed memory system The corresponding localization node in family end, the complete copy data for storing the described first data to be written；
Wherein, step 23 is specially：Obtain the first segment for the distributed memory system that the client is currently connected Point, is used as localization node；In the embodiment, the node currently connected according to client is as localization node.
Or, step 23 is specially：It is that the client specifies one first in the distributed memory system interior joint Node, is used as localization node.In the embodiment, a node is specified as localization node, subsequently can be by local side Node address is sent to client so that client is directly connected to localize node.The step is specially：According to the distribution The performance of the current appearance load rate of the node of storage system, each node in the distributed memory system, and/or connected visitor The quantity at family end, in the distributed memory system interior joint, is that the client specifies a first node, is used as localization Node.Localization node is specified according to above-mentioned strategy, it is contemplated that the factor of each side, is more beneficial for the stabilization of system and is Equilibrium between system.
Step 24, it is that the client selects at least one node as the corresponding strange land node of the client, is used for The complete copy data of the predetermined quantity of distribution storage first data to be written；
Step 24 is specifically, according to current appearance load rate, the distributed storage of the node of the distributed memory system The performance of each node in system, is that client selects at least one Section Point, is used as the corresponding strange land of the client Node.According to the specified strange land node of above-mentioned strategy, it is contemplated that the factor of each side, is more beneficial for the stabilization and system of system Between equilibrium.
Step 25, it will be localized described in a complete copy data Cun Chudao of the described first data to be written in node；Will The complete copy distributed data storage of the predetermined quantity of first data to be written is into the localization node.
It is that client creates a volume in distributed memory system, one of the volume is complete in the embodiment of the present invention Copy data set on one node, when client access distributed memory system volume when, can connect and preserve volume Complete copy data localization node, so as to accelerate access speed of the client to volume.
Optionally, methods described also includes：
Step 26, it is each in the distributed memory system according to all copy datas of the described first data to be written The distribution of individual node, records the storage address letter of each node of the copy data of the volume in the distributed memory system Breath.
Methods described also includes：
Step 27, when the copy data operational order to the described first data to be written for receiving the client transmission When, according to the operational order, the copy data to the described first data to be written in the strange land node is operated；
Step 28, the copy data according to the operational order to the described first data to be written in the strange land node enters After row operation, the copy data according to the operational order to the described first data to be written in the localization node is grasped Make.
Optionally, between the step 21 and step 22, methods described also includes：
In the localization node, volume controller is created, the volume controller is used for the pair for recording the localization volume The storage address information of node of the notebook data in the distributed memory system, and for controlling to read the copy data Data consistency and control the IO orders of the copy data；
Accordingly, the step 26 is specially：Volume controller is according to all copy datas of the described first data to be written The distribution of each node in the distributed memory system, records the copy data of the volume in the distributed storage system The storage address information of each node in system.
As shown in figure 3, be a kind of processing unit to being rolled up in distributed memory system of the present invention, including：
Creating unit 31, is that client creates a volume, the physics space-consuming of the volume in distributed memory system For the predefined size of the volume；
Selecting unit 32, is that client selects a first node as the visitor in the node of distributed memory system The corresponding localization node in family end；It is that client selects at least one Section Point to make in the node of distributed memory system For the corresponding strange land node of the client；
First allocation unit 33, is the first address that the client distributes predefined size in the localization node, A complete copy data for storing the volume；
Second allocation unit 34, is the ground that the volume distributes the predefined size at least one described strange land node Location, the complete copy data of the predetermined quantity for being distributed the storage volume；
Receiving unit 35, receives the client to first write data requests permed and sent, described first writes data Request includes the first data to be written；
First writing unit 36, by a complete copy data of the described first data to be written, writes the first node In；
Second writing unit 37, by the complete copy data of the predetermined quantity of the described first data to be written, distribution write-in In the strange land node.
As shown in figure 4, be a kind of operation device to being rolled up in distributed memory system of the present invention, including：
Creating unit 41, is that client creates a volume, the virtual space-consuming of the volume in distributed memory system For the predefined size of the volume, the initial physical space-consuming of the volume is 0；
Receiving unit 42, receives the client to first write data requests permed and sent, described first writes data Request includes the first data to be written；
First choice unit 43, is that the client selects a first node to make in the node of distributed memory system For the corresponding localization node of the client, the complete copy data for storing the described first data to be written；
Second selecting unit 44, is that the client selects at least one node to be saved as the corresponding strange land of the client Point, the complete copy data of the predetermined quantity for being distributed storage first data to be written；
Memory cell 45, will localize node described in a complete copy data Cun Chudao of the described first data to be written In；By the complete copy distributed data storage of the predetermined quantity of the described first data to be written into the localization node.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto, Any one skilled in the art the invention discloses technical scope in, the change or replacement that can be readily occurred in, It should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be with scope of the claims It is defined.

Claims (10)

1. a kind of processing method to being rolled up in distributed memory system, it is characterised in that including：
Step 1, it is that client creates a volume, the physics space-consuming of the volume is the volume in distributed memory system Predefined size；
Step 2, it is that client selects a first node corresponding as the client in the node of distributed memory system Localize node；
Step 3, it is that client selects at least one Section Point as the client in the node of distributed memory system Corresponding strange land node；
Step 4, it is the first address that the client distributes predefined size in the localization node, it is described for storing One complete copy data of volume；
Step 5, at least one described strange land node, it is the address that the volume distributes the predefined size, is deposited for being distributed Store up the complete copy data of the predetermined quantity of the volume；
Step 6, the client is received to first write data requests permed and sent, and first write data requests include the One data to be written；
Step 7, by a complete copy data of the described first data to be written, write in the first node；
Step 8, by the complete copy data of the predetermined quantity of the described first data to be written, the distribution write-in strange land node In.
2. according to the method described in claim 1, it is characterised in that methods described also includes：
When receiving that the client sends to the operational orders of the described first data to be written, according to the operational order, Copy data to the described first data to be written in the strange land node is operated according to the operational order；
When the copy data in the strange land node of the described first data to be written carries out having operated according to the operational order Cheng Hou, the copy data according to the operational order to the described first data to be written in the localization node is operated.
3. according to the method described in claim 1, it is characterised in that described in the node of distributed memory system, select one The step of first node is as localization node includes：
The first node for the distributed memory system that the client is currently connected is obtained, localization node is used as；Or
In the distributed memory system interior joint, it is that the client specifies a first node, is used as localization node.
4. method according to claim 3, it is characterised in that described in the distributed memory system interior joint, is The client specifies a first node, includes as the step of localizing node：
Each node in the current appearance load rate of the node of the distributed memory system, the distributed memory system Performance, and/or the quantity of client has been connected, it is that the client specifies one in the distributed memory system interior joint First node, is used as localization node.
5. it is according to the method described in claim 1, it is characterised in that described in the node of distributed memory system, client The step of end selects at least one Section Point corresponding as client strange land node includes：
Each node in the current appearance load rate of the node of the distributed memory system, the distributed memory system Performance, is that client selects at least one Section Point, is used as the corresponding strange land node of the client.
6. a kind of processing method to being rolled up in distributed memory system, it is characterised in that including：
Step 1, it is that client creates a volume, the virtual space-consuming of the volume is the volume in distributed memory system Predefined size, the initial physical space-consuming of the volume is 0；
Step 2, the client is received to first write data requests permed and sent, and first write data requests include the One data to be written；
Step 3, it is that the client selects a first node as the client pair in the node of distributed memory system The localization node answered, the complete copy data for storing the described first data to be written；
Step 4, it is that the client selects at least one node as the corresponding strange land node of the client, is deposited for being distributed Store up the complete copy data of the predetermined quantity of the described first data to be written；
Step 5, it will be localized described in a complete copy data Cun Chudao of the described first data to be written in node；By described The complete copy distributed data storage of the predetermined quantity of one data to be written is into the localization node.
7. method according to claim 6, it is characterised in that methods described also includes：
When receiving that the client sends to the copy data operational orders of the described first data to be written, according to the behaviour Instruct, the copy data to the described first data to be written in the strange land node is operated；
According to the operational order to the described first data to be written after the copy data of the strange land node is operated, root Copy data according to the operational order to the described first data to be written in the localization node is operated.
8. method according to claim 6, it is characterised in that described in the node of distributed memory system, is described Client selects a step of first node is as localization node to include：
The first node for the distributed memory system that the client is currently connected is obtained, localization node is used as；Or
In the distributed memory system interior joint, it is that the client specifies a first node, is used as localization node.
9. a kind of processing unit to being rolled up in distributed memory system, it is characterised in that including：
Creating unit, is that client creates a volume, the physics space-consuming of the volume is described in distributed memory system The predefined size of volume；
Selecting unit, is that client selects a first node as the client pair in the node of distributed memory system The localization node answered；It is that client selects at least one Section Point as described in the node of distributed memory system The corresponding strange land node of client；
First allocation unit, is the first address that the client distributes predefined size, for depositing in the localization node Store up a complete copy data of the volume；
Second allocation unit, at least one described strange land node, is the address that the volume distributes the predefined size, is used for The complete copy data of the predetermined quantity of the distribution storage volume；
Receiving unit, receives the client to first write data requests permed and sent, the first write data requests bag Include the first data to be written；
First writing unit, a complete copy data of the described first data to be written are write in the first node；
Second writing unit, by the complete copy data of the predetermined quantity of the described first data to be written, distribution write-in is described different In ground node.
10. a kind of operation device to being rolled up in distributed memory system, it is characterised in that including：
Creating unit, is that client creates a volume, the virtual space-consuming of the volume is described in distributed memory system The predefined size of volume, the initial physical space-consuming of the volume is 0；
Receiving unit, receives the client to first write data requests permed and sent, the first write data requests bag Include the first data to be written；
First choice unit, is that the client selects a first node as described in the node of distributed memory system The corresponding localization node of client, the complete copy data for storing the described first data to be written；
Second selecting unit, is that the client selects at least one node as the corresponding strange land node of the client, uses In the complete copy data of the predetermined quantity of distribution storage first data to be written；
Memory cell, will be localized in node described in a complete copy data Cun Chudao of the described first data to be written；By institute The complete copy distributed data storage of predetermined quantity of the first data to be written is stated into the localization node.
CN201710213932.8A 2017-04-01 2017-04-01 To the treating method and apparatus rolled up in distributed memory system Pending CN107037987A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710213932.8A CN107037987A (en) 2017-04-01 2017-04-01 To the treating method and apparatus rolled up in distributed memory system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710213932.8A CN107037987A (en) 2017-04-01 2017-04-01 To the treating method and apparatus rolled up in distributed memory system

Publications (1)

Publication Number Publication Date
CN107037987A true CN107037987A (en) 2017-08-11

Family

ID=59533972

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710213932.8A Pending CN107037987A (en) 2017-04-01 2017-04-01 To the treating method and apparatus rolled up in distributed memory system

Country Status (1)

Country Link
CN (1) CN107037987A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111163120A (en) * 2018-11-08 2020-05-15 阿里巴巴集团控股有限公司 Data storage and transmission method and device of distributed database and storage medium
CN112395281A (en) * 2020-12-10 2021-02-23 太极计算机股份有限公司 Heterogeneous multi-source data fusion system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101741911A (en) * 2009-12-18 2010-06-16 中兴通讯股份有限公司 Multi-copy collaboration-based write operation method, system and node
CN102546782A (en) * 2011-12-28 2012-07-04 北京奇虎科技有限公司 Distribution system and data operation method thereof
US9477565B2 (en) * 2013-01-22 2016-10-25 Tencent Technology (Shenzhen) Company Limited Data access with tolerance of disk fault
CN106484311A (en) * 2015-08-31 2017-03-08 华为数字技术（成都）有限公司 A kind of data processing method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101741911A (en) * 2009-12-18 2010-06-16 中兴通讯股份有限公司 Multi-copy collaboration-based write operation method, system and node
CN102546782A (en) * 2011-12-28 2012-07-04 北京奇虎科技有限公司 Distribution system and data operation method thereof
US9477565B2 (en) * 2013-01-22 2016-10-25 Tencent Technology (Shenzhen) Company Limited Data access with tolerance of disk fault
CN106484311A (en) * 2015-08-31 2017-03-08 华为数字技术（成都）有限公司 A kind of data processing method and device

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111163120A (en) * 2018-11-08 2020-05-15 阿里巴巴集团控股有限公司 Data storage and transmission method and device of distributed database and storage medium
CN112395281A (en) * 2020-12-10 2021-02-23 太极计算机股份有限公司 Heterogeneous multi-source data fusion system
CN112395281B (en) * 2020-12-10 2021-05-11 太极计算机股份有限公司 Heterogeneous multi-source data fusion system

Similar Documents

Publication Publication Date Title
US8019965B2 (en) Data migration
US8725946B2 (en) Mass storage system and method of using hard disk, solid-state media, PCIe edge connector, and raid controller
CN102789367B (en) Dynamic area
US20120102261A1 (en) Systems and Methods for Tiered Non-Volatile Storage
CN104516471B (en) Method and device for managing power supply of storage system
DE102017104150A1 (en) Wear compensation in storage devices
CN104503703B (en) The treating method and apparatus of caching
TW200917022A (en) Control method based on multi-passage flash memory apparatus logic strip
CN107665096A (en) Weighted data slitting
CN104778018A (en) Broad-strip disk array based on asymmetric hybrid type disk image and storage method of broad-strip disk array
CN104461384B (en) A kind of method for writing data and storage device
CN106708433A (en) Read-write method and device for stored data
CN108170614A (en) A kind of method, apparatus that Data duplication is avoided to recycle and solid state disk
CN106326384A (en) File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array)
WO2001073537A2 (en) System for and method of accessing blocks on a storage medium
CN104834478B (en) A kind of data write-in and read method based on isomery mixing storage device
CN107037987A (en) To the treating method and apparatus rolled up in distributed memory system
CN107145301B (en) Data access method and device based on hybrid disk
CN107092443B (en) Data migration method and device
CN101997919B (en) Storage resource management method and device
CN105912279B (en) Solid-state storage recovery system and solid-state storage recovery method
US20130086317A1 (en) Passing hint of page allocation of thin provisioning with multiple virtual volumes fit to parallel data access
CN106919342A (en) Storage resource distribution method and device based on automatic simplify configuration
US9588898B1 (en) Fullness control for media-based cache operating in a steady state
KR20110054788A (en) Apparatus and method for storing data using non-volatile buffer

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170811