CN107463342A - A kind of storage method and device of CDN fringe nodes file - Google Patents
A kind of storage method and device of CDN fringe nodes file Download PDFInfo
- Publication number
- CN107463342A CN107463342A CN201710752436.XA CN201710752436A CN107463342A CN 107463342 A CN107463342 A CN 107463342A CN 201710752436 A CN201710752436 A CN 201710752436A CN 107463342 A CN107463342 A CN 107463342A
- Authority
- CN
- China
- Prior art keywords
- hard disk
- file
- stored
- cryptographic hash
- hash
- 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
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
- G06F3/0614—Improving the reliability of storage systems
-
- 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
- G06F3/0614—Improving the reliability of storage systems
- G06F3/0617—Improving the reliability of storage systems in relation to availability
-
- 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
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0643—Management of files
-
- 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)
- Storage Device Security (AREA)
Abstract
The storage method and device for the CDN fringe node files that example of the present invention provides, from the raid card for supporting Nonraid patterns, obtain the hard disk name of each hard disk；And it is connected by the hard disk groups with edge node server；The hard disk groups are initialized；According to the hard disk name of each hard disk and the cryptographic Hash and preset rules of the filename of each file to be stored, each file to be stored is stored into the corresponding hard disk respectively.Every piece of hard disk is in the state of independent operating, and avoiding establishment raid5 in the prior art needs one piece of hard disk to make redundancy disk, improves the maximum capacity of edge node server；When even if there is two pieces and more than two pieces of faulty disk, will not also cause whole server unavailable, and change faulty disk, only need the re-synchronization edge node server to belong to the data of the faulty disk, fault time is greatly shortened, improve the stability of system.
Description
Technical field
The present invention relates to file storage technology field, more particularly to a kind of storage method of CDN fringe nodes file
And device.
Background technology
CDN full name is Content Delivery Network, i.e. content distributing network.Its basic ideas is as far as possible
Avoid being possible to influenceing the bottleneck and link of data transmission bauds and stability on internet, make content transmission faster, it is more steady
It is fixed.By placing node server everywhere in network, one layer of intelligent virtual network being formed on existing Internet basic,
CDN system can in real time according to the connection of network traffics and each node, load state and distance and the response to user when
Between etc. integrated information the request of user is re-directed on the service node nearest from user.The purpose is to user is taken nearby
Content needed for obtaining, solve the crowded situation of Internet network, improve the response speed that user accesses website.
CDN is because its superiority is widely used in technical field of video storage, and video CD N is mainly by Centroid and side
Edge node two parts are formed.In video CDN, Centroid is mainly made up of GSLB device, and overall situation load is equal
Fringe node address currently closest to user is supplied to user by weighing apparatus by the strategy pre-defined.It is meanwhile global negative
Carry balanced device also with other fringe nodes keep communicate, collect each nodal information, it is ensured that user will not be dispatched to it is unavailable or
Unsound fringe node.Fringe node is a kind of cache server nearest from user.
Controlled by GSLB device, the request of user is pointed at his nearest node, and the high speed in node is delayed
Server is deposited just as being in same LAN with user, the request of user can be responded faster.In order to faster more fully ring
Using family request, it is necessary to edge node server capacity more greatly and stably.
At present in fringe node file storage technology, all sata data disks and ssd data disks are set up one respectively
Raid5, data are stored entirely in the raid5 of sata disks establishment；Then all ssd data disks are set up into a raid5, data
It is stored entirely in the raid5 of ssd disks establishment.
In above method, technological deficiency be present：After raid5, when one piece of big capacity hard disk breaks down, replacement is former
After hindering disk, rebuild was for up to several days, if continuing that hard disk failure occurs during rebuild, whole raid5 data all will
Lose, the problem of causing the fault recovery cycle long.
The content of the invention
In view of this, the invention provides a kind of storage method and device of CDN fringe nodes file, file distribution to deposit
Store up in each hard disk in fringe node, make fringe node maximum capacity, improve edge node server and faulty disk be present
Service ability afterwards and failover capability.
Concrete technical scheme is as follows：
A kind of storage method of CDN fringe nodes file, including：
The raid card for supporting Nonraid patterns is obtained, and hard disk groups and edge node server are made by the raid card
It is connected；
The hard disk groups are initialized, obtain the hard disk name of each hard disk in the hard disk groups；
The hard disk name of each hard disk and each file to be stored in the hard disk groups are calculated according to uniformity hash algorithm
The cryptographic Hash of filename, and according to the hard disk name and the cryptographic Hash of the filename of each file to be stored of each hard disk
And preset rules, each file to be stored is stored into the corresponding hard disk respectively.
Preferably, it is described that the hard disk groups are initialized, the hard disk name of each hard disk in the hard disk groups is obtained, is wrapped
Include：
Each hard disk in the hard disk groups is identified, determines the equipment symbol title and class of each hard disk respectively
Type, the hard disk type are sata hard disks or ssd hard disks；
Subregion is formatted to each hard disk respectively；
According to the quantity of hard disk in the hard disk groups, hard disk serial is established；
According to the traversal order to each hard disk, each hard disk is mounted in the hard disk serial respectively,
And the hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
Preferably, it is described to calculate the hard disk name of each hard disk in the hard disk groups according to uniformity hash algorithm and each treat
The cryptographic Hash of the filename of storage file, and according to the hard disk name of each hard disk and the file of each file to be stored
The cryptographic Hash and preset rules of name, each file to be stored is stored into the corresponding hard disk respectively, including：
According to uniformity hash algorithm, to each memory space less than the hard disk name of the hard disk carry out Hash, calculate
Go out each memory space less than the hard disk hard disk name cryptographic Hash, according to each memory space less than the hard disk
The cryptographic Hash of hard disk name, each memory space less than the hard disk form a Hash ring, obtain each memory space less than
Position of the hard disk on the Hash ring；
According to uniformity hash algorithm, Hash is carried out to the filename of each file to be stored, calculates each institute
State the cryptographic Hash of the filename of file to be stored；
According to the filename of each file to be stored and the cryptographic Hash of the hard disk name of each hard disk, treated each
Storage file store into the corresponding hard disk on the Hash ring, make each file to be stored filename cryptographic Hash with
The difference of the cryptographic Hash of the hard disk name of respective stored hard disk is minimum, and the cryptographic Hash of the filename of each file to be stored is small
In the cryptographic Hash of the hard disk name of corresponding storage hard disk.
Preferably, the hard disk groups are sata hard disk groups, and each file to be stored storage is arrived into phase respectively described
After in the hard disk answered, methods described also includes：
In preset time period, it is determined that multiple focus files in the file stored in sata hard disk groups；
Calculated according to uniformity hash algorithm every in the cryptographic Hash and ssd hard disk groups of the filename of each focus file
The cryptographic Hash of the hard disk name of individual ssd hard disks, and the cryptographic Hash and each ssd of the filename according to each focus file
The cryptographic Hash and preset rules of the hard disk name of hard disk, it is hard that each focus file is dumped into the corresponding ssd respectively
In disk.
Preferably, it is described in preset time period, it is determined that multiple focuses in the file stored in sata hard disk groups
File, including：
In preset time period, the visitation frequency of each file stored in the sata hard disk groups is counted；
The each file stored in the sata hard disk groups is ranked up from high to low by visitation frequency；
The file of the high visitation frequency of preset ratio in the sequence is defined as focus file.
Preferably, the cryptographic Hash of the filename that each focus file is calculated according to uniformity hash algorithm and
The cryptographic Hash of the hard disk name of each ssd hard disks in ssd hard disk groups, and the cryptographic Hash of the filename according to each focus file
With the cryptographic Hash and preset rules of the hard disk name of each ssd hard disks, each focus file is dumped into phase respectively
In the ssd hard disks answered, including：
According to uniformity hash algorithm, to each memory space less than the hard disk names of the ssd hard disks carry out Hash, meter
Calculate each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each memory space less than described in
The cryptographic Hash of the hard disk name of ssd hard disks, each memory space less than the ssd hard disks form a Hash ring, obtain each
Memory space less than position of the ssd hard disks on the Hash ring；
According to uniformity hash algorithm, Hash is carried out to the filename of each focus file, calculated each described
The cryptographic Hash of the filename of focus file；
, will be each described according to the filename of each focus file and the cryptographic Hash of the hard disk name of each ssd hard disks
Focus file is dumped in the corresponding ssd hard disks on the Hash ring, makes the cryptographic Hash of the filename of each focus file
It is minimum with the difference of the cryptographic Hash of the hard disk name of respective stored ssd hard disks, and the Hash of the filename of each focus file
Value is less than the cryptographic Hash of the hard disk name of corresponding storage ssd hard disks.
Preferably, methods described also includes：
When in corresponding Hash ring be not present more than file to be stored the cryptographic Hash of filename or the text of focus file
During the hard disk of the cryptographic Hash of part name, the starting that the file to be stored or the focus file are stored in corresponding Hash ring is hard
On disk.
A kind of storage device of CDN fringe nodes file, including：
Acquiring unit, the raid card of Nonraid patterns is supported for obtaining, and makes hard disk groups and side by the raid card
Edge node server is connected；
Initialization unit, for being initialized to the hard disk groups, obtain the hard disk of each hard disk in the hard disk groups
Name；
First memory cell, for according to uniformity hash algorithm calculate in the hard disk groups hard disk name of each hard disk and
The cryptographic Hash of the filename of each file to be stored, and the hard disk name according to each hard disk and each file to be stored
Filename cryptographic Hash and preset rules, respectively will each file to be stored storage into the corresponding hard disk.
Preferably, the initialization unit includes：
Subelement is identified, for each hard disk in the hard disk groups to be identified, determines each hard disk respectively
Equipment symbol title and type, the hard disk type be sata hard disks or ssd hard disks；
Subregion subelement is formatted, for being formatted subregion to each hard disk respectively；
Subelement is established, for the quantity according to hard disk in the hard disk groups, establishes hard disk serial；
Carry subelement, for according to the traversal order to each hard disk, being respectively mounted to each hard disk
In the hard disk serial, and the hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
Preferably, first memory cell includes：
First computation subunit, for according to uniformity hash algorithm, to each memory space less than the hard disk
Hard disk name carry out Hash, calculate each memory space less than the hard disk hard disk name cryptographic Hash, according to each storage
Space less than the hard disk hard disk name cryptographic Hash, each memory space less than the hard disk form a Hash ring,
Obtain each memory space less than position of the hard disk on the Hash ring；
Second computation subunit, for according to uniformity hash algorithm, entering to the filename of each file to be stored
Row Hash, calculate the cryptographic Hash of the filename of each file to be stored；
First storing sub-units, the hard disk for the filename according to each file to be stored and each hard disk
The cryptographic Hash of name, by the storage of each file to be stored into the corresponding hard disk on the Hash ring, make each text to be stored
The difference of the cryptographic Hash of the filename of part and the cryptographic Hash of the hard disk name of respective stored hard disk is minimum, and each text to be stored
The cryptographic Hash of the filename of part is less than the cryptographic Hash of the hard disk name of corresponding storage hard disk.
Preferably, the hard disk groups are sata hard disk groups, and described device also includes：
Determining unit, in preset time period, it is determined that multiple heat in the file stored in sata hard disk groups
Dot file；
Second memory cell, the Hash of the filename for calculating each focus file according to uniformity hash algorithm
The cryptographic Hash of the hard disk name of each ssd hard disks in value and ssd hard disk groups, and the Kazakhstan of the filename according to each focus file
The cryptographic Hash and preset rules of uncommon value and the hard disk name of each ssd hard disks, respectively will each focus file unloading
Into the corresponding ssd hard disks.
Preferably, the determining unit includes：
Subelement is counted, in preset time period, counting the visit of each file stored in the sata hard disk groups
Ask the frequency；
Sort subelement, for being carried out from high to low to each file stored in the sata hard disk groups by visitation frequency
Sequence；
Determination subelement, for the file of the high visitation frequency of preset ratio in the sequence to be defined as into focus file.
Preferably, second memory cell includes：
3rd computation subunit, for according to uniformity hash algorithm, to each memory space less than the ssd hard disks
Hard disk name carry out Hash, calculate each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each
Memory space less than the ssd hard disks hard disk name cryptographic Hash, each memory space less than the ssd hard disks form one
Individual Hash ring, obtain each memory space less than position of the ssd hard disks on the Hash ring；
4th computation subunit, for according to uniformity hash algorithm, being carried out to the filename of each focus file
Hash, calculate the cryptographic Hash of the filename of each focus file；
Second storing sub-units, the hard disk name for filename and each ssd hard disks according to each focus file
Cryptographic Hash, each focus file is dumped in the corresponding ssd hard disks on the Hash ring, makes each focus text
The difference of the cryptographic Hash of the filename of part and the cryptographic Hash of the hard disk name of respective stored ssd hard disks is minimum, and each focus
The cryptographic Hash of the filename of file is less than the cryptographic Hash of the hard disk name of corresponding storage ssd hard disks.
Preferably, described device also includes：
3rd memory cell, for when the Hash that the filename more than file to be stored is not present in corresponding Hash ring
During the hard disk of the cryptographic Hash of the filename of value or focus file, the file to be stored or the focus file are stored in accordingly
On starting hard disk in Hash ring.
Preferably, described device also includes：
3rd memory cell, for when the Hash that the filename more than file to be stored is not present in corresponding Hash ring
During the hard disk of the cryptographic Hash of the filename of value or focus file, the file to be stored or the focus file are stored in accordingly
On starting hard disk in Hash ring.
It is as follows relative to prior art, beneficial effects of the present invention：
The storage method and device of CDN fringe nodes file provided by the invention, from the raid for supporting Nonraid patterns
Card, every piece of hard disk are in the state of independent operating, and avoiding establishment raid5 in the prior art needs one piece of hard disk to make redundancy disk,
The maximum capacity of edge node server is improved, covers more files, source request is reduced back, improves Consumer's Experience；
The present invention is according to the hard disk name of each hard disk and the cryptographic Hash of the filename of each file to be stored and default rule
Then, each file to be stored is stored into the corresponding hard disk respectively, ensures that file to be stored is uniformly stored in
In each hard disk, due to not needing one piece of hard disk to make redundancy disk, each hard disk is individually present and storage file, even if occurring two pieces
And more than two pieces of faulty disk, when will not also cause whole server unavailable, and change faulty disk, only need the re-synchronization side
Edge node server belongs to the data of the faulty disk, and fault time is greatly shortened, and improves the stability of system.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of storage method flow chart of CDN fringe nodes file disclosed in the embodiment of the present invention；
Fig. 2 is the storage method flow chart of another CDN fringe nodes file disclosed in the embodiment of the present invention；
Fig. 3 is the storage method flow chart of another CDN fringe node file disclosed in the embodiment of the present invention；
Fig. 4 is the storage method flow chart of another CDN fringe node file disclosed in the embodiment of the present invention；
Fig. 5 is a kind of memory device structure schematic diagram of CDN fringe nodes file disclosed in the embodiment of the present invention；
Fig. 6 is the memory device structure schematic diagram of another CDN fringe nodes file disclosed in the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Referring to Fig. 1, present embodiment discloses a kind of storage method of CDN fringe nodes file, following step is specifically included
Suddenly：
S101：The raid card for supporting Nonraid patterns is obtained, and hard disk groups is taken with fringe node by the raid card
Business device is connected；
Nonraid patterns are that every piece of hard disk is a Logical Disk, does not have data block slitting without using raid patterns.
From the raid card for supporting Nonraid patterns, every piece of hard disk is in independent operating state, avoided in the prior art
Setting up raid5 needs one piece of hard disk to make redundancy disk, all hard disks can storage file, improve edge node server most
Large Copacity, the effective document amount of edge node server is added, covers more files, when being sent out in edge node server
During raw hard disk failure, scheduling system carries out fault redundance, and the request of script access fault disk is dispatched to other fringe nodes, subtracted
Source request is returned less, improves Consumer's Experience.
Hard disk in edge node server uniformly connects raid card by data wire, raid card again with fringe node service
The mainboard of device is connected.
Preferably, the hard disk groups are sata hard disk groups.
Sata hard disk groups include multiple sata hard disks, and the hard disk of sata (Serial ATA) mouth is called serial port hard disk, sata
Hard disk uses serial connection mode, and Serial ATA bus uses embedded clock signal, possesses stronger error correcting capability, with
Past hard disk compared to its maximum difference be can to transmission instruction check, if it find that mistake can automatic straightening, this
Greatly enhance the reliability of data transfer.Serial line interface also has the advantages of simple in construction, support hot plug.
S102：The hard disk groups are initialized, obtain the hard disk name of each hard disk in the hard disk groups；
Specifically, S102 implementation procedure is as follows：
Each hard disk in the hard disk groups is identified, determines the equipment symbol title and class of each hard disk respectively
Type, the hard disk type are sata hard disks or ssd hard disks；
Equipment symbol title characterizes the mark of each hard disk, the equipment under windows be with " Device [equipment
Name] " form name.Such as：The c disks of disk partition, d disks equipment symbol title be respectively just " Device
HarddiskVolume1"、"\Device\HarddiskVolume2"。
Subregion is formatted to each hard disk respectively；
Format a kind of operation that subregion refers to initialize the subregion (partition) in hard disk, this operation
Cause file all in existing hard disk to be eliminated, make storage system more stable.It is that hard disk groups are carried out to format subregion
The steps necessary of initialization, distinguished without formatting, the hard disk in hard disk groups can not be used.
According to the quantity of hard disk in the hard disk groups, hard disk serial is established；
According to each hard disk traversal order, each hard disk is mounted in the hard disk serial respectively, and
Hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
Specifically, it is sata hard disks or ssd hard disks by smartctl command recognition hard disks.
Assuming that have 12 pieces of hard disks, hard disk serial data1, data2 ..., data12.
Each hard disk has respective equipment symbol title in the server, is calculated for the ease of cryptographic Hash, according to each
Time the traversal order of the hard disk, the mapping relations of sequential value in equipment symbol title and the hard disk serial of each hard disk are established, i.e.,
Sequential value corresponding to first hard disk gone through is data1, and sequential value corresponding to second hard disk of traversal is data2, with such
Push away.Each hard disk is mounted in the hard disk serial respectively, that is to say, that after carry, each hard disk is corresponding
One sequential value, and using sequential value as hard disk name, for example, the entitled data1 of hard disk of first hard disk of traversal.
The present embodiment auto-initiation hard disk, equipment symbol title, the type of automatic identification hard disk, and the number per class hard disk
Amount, the manually configuration raid processes that will be based on become to find automatically and complete to initialize, save manpower and time, improve
The flexibility of initialization operation.
S103：The hard disk name of each hard disk and each text to be stored in the hard disk groups are calculated according to uniformity hash algorithm
The cryptographic Hash of the filename of part, and the hard disk name according to each hard disk and the Kazakhstan of the filename of each file to be stored
Uncommon value and preset rules, each file to be stored is stored into the corresponding hard disk respectively.
Specifically, referring to Fig. 2, S103 implementation procedure is as follows：
S201：According to uniformity hash algorithm, to each memory space less than the hard disk name of the hard disk carry out Hash,
Calculate each memory space less than the hard disk hard disk name cryptographic Hash, according to each memory space less than it is described hard
The cryptographic Hash of the hard disk name of disk, each memory space less than the hard disk form a Hash ring, obtain each memory space
Less than position of the hard disk on the Hash ring；
S202：According to uniformity hash algorithm, Hash is carried out to the filename of each file to be stored, calculated every
The cryptographic Hash of the filename of the individual file to be stored；
S203：, will according to the filename of each file to be stored and the cryptographic Hash of the hard disk name of each hard disk
Each file to be stored storage makes the Kazakhstan of the filename of each file to be stored into the corresponding hard disk on the Hash ring
The difference of uncommon value and the cryptographic Hash of the hard disk name of respective stored hard disk is minimum, and the Kazakhstan of the filename of each file to be stored
Uncommon value is less than the cryptographic Hash of the hard disk name of corresponding storage hard disk.
It is understood that multiple files can be stored on each hard disk, wherein, file can be any type of file,
For example, audio file, video file etc..Until disk space is zero, that is to say, that in storage file, only store file
Memory space less than hard disk in.
Each hard disk has respective hard disk name, for example, data1, data2, etc., each file to be stored also has each
Pre-set filename, for example, file1, file2, etc..
According to uniformity hash algorithm, to each memory space less than the hard disk name of the hard disk carry out Hash, calculate
Go out each memory space less than the hard disk hard disk name cryptographic Hash, for example, data1 cryptographic Hash is node_
Hashcode, according to each memory space less than the hard disk hard disk name cryptographic Hash, each memory space less than institute
State hard disk and form a Hash ring, obtain each memory space less than position of the hard disk on the Hash ring, then,
According to uniformity hash algorithm, Hash is carried out to the filename of each file to be stored, calculated each described to be stored
The cryptographic Hash of the filename of file, such as file1 cryptographic Hash is key_hashcode.So, the hard disk name of each hard disk
Cryptographic Hash and the value that the cryptographic Hash of the filename of each file to be stored is same type, being easy to will be to be stored according to preset rules
File is uniformly stored in each hard disk.
According to the cryptographic Hash and preset rules of each hard disk and each file to be stored, respectively by each institute
File to be stored storage is stated into the corresponding hard disk.Specifically, the preset rules are：
The difference of the cryptographic Hash and the cryptographic Hash of the hard disk name of respective stored hard disk of the filename of each file to be stored
Value is minimum, and the cryptographic Hash of the filename of each file to be stored is less than the cryptographic Hash of the hard disk name of corresponding storage hard disk.
Such as：When the cryptographic Hash of the filename of each file to be stored is respectively 1,5,9, the Kazakhstan of the hard disk name of each hard disk
Uncommon value is respectively 2,6,10.File to be stored corresponding to 1 is just stored in hard disk corresponding to 2 by that, file to be stored corresponding to 5
It is stored in hard disk corresponding to 6, file to be stored is stored in hard disk corresponding to 10 corresponding to 9.
The cryptographic Hash of the filename of the file to be stored is less than the cryptographic Hash of the hard disk name of the hard disk of corresponding storage location
Purpose be to ensure clockwise storage of the file to be stored on Hash ring, avoid having more than one file to be stored simultaneously
The difference of the cryptographic Hash of the cryptographic Hash of filename and the hard disk name of same hard disk is all minimum, and the file of a file to be stored
The cryptographic Hash of name is more than the cryptographic Hash of the hard disk name of the hard disk, and the cryptographic Hash of the filename of another file to be stored is hard less than this
During the cryptographic Hash of the hard disk name of disk, two files all store the situation of the hard disk, ensure the uniform storage of file to be stored.
Key_hashcode≤node_hashcode is not present it should be noted that working as, then, corresponding to file f ile1
Storage hard disk is data1, and data1 is the starting hard disk in Hash ring.
The present embodiment provide CDN fringe node files storage method, from support Nonraid patterns raid card,
Every piece of hard disk is in the state of independent operating, and avoiding establishment raid5 in the prior art needs one piece of hard disk to make redundancy disk, improves
The maximum capacity of edge node server, covers more files, reduces back source request, improves Consumer's Experience；This reality
Hard disk name and the cryptographic Hash and preset rules of the filename of each file to be stored of the example according to each hard disk are applied,
Each file to be stored is stored into the corresponding hard disk respectively, it is each to ensure that file to be stored is uniformly stored in
In hard disk, because the present embodiment does not need one piece of hard disk to make redundancy disk, each hard disk is individually present and storage file, even if occurring
When more than two pieces of faulty disk, will not also cause whole server unavailable, and change faulty disk, the re-synchronization edge is only needed
Node server belongs to the data of the faulty disk, and fault time is greatly shortened, and improves the stability of system.
Ssd hard disk groups include multiple ssd hard disks, ssd hard disks (Solid State Drives, solid state hard disc) solid-state electricity
Sub- storage chip array and manufactured hard disk, are made up of control unit and memory cell (FLASH chip, dram chip).Ssd is hard
The performance of disk is far above sata hard disks.
Because the performance of ssd hard disks is far above sata hard disks, in order to reduce the pressure of sata hard disks, ssd hard disks are improved
Utilization rate, the higher focus file of access degree is dumped in ssd hard disks.On the basis of embodiment corresponding to Fig. 1, refer to
Fig. 3, present embodiment discloses the storage method of another CDN fringe nodes file, specifically include：
S301：The raid card for supporting Nonraid patterns is obtained, and makes sata hard disk groups and ssd hard by the raid card
Disk group is connected with edge node server；
S302：The sata hard disk groups and the ssd hard disk groups are initialized, obtain the sata hard disk groups and institute
State the hard disk name of each hard disk in ssd hard disk groups；
S303：The hard disk name of each sata hard disks and each file to be stored are calculated according to uniformity hash algorithm
The cryptographic Hash of filename, and the hard disk name according to each sata hard disks and the Kazakhstan of the filename of each file to be stored
Uncommon value and preset rules, each file to be stored is stored into the corresponding sata hard disks respectively；
S304：In preset time period, it is determined that multiple focuses text in the file stored in the sata hard disk groups
Part；
Specifically, S304 implementation procedure is as follows：
In preset time period, the visitation frequency of each file stored in the sata hard disk groups is counted；
The each file stored in the sata hard disk groups is ranked up from high to low by visitation frequency；
The file of the high visitation frequency of preset ratio in the sequence is defined as focus file.
It should be noted that preset time period and preset ratio, can be set in advance according to actual conditions, for example,
Preset time period can be 1 hour, and preset ratio can be 50%.
S305：The cryptographic Hash and ssd hard disks of the filename of each focus file are calculated according to uniformity hash algorithm
The cryptographic Hash of the hard disk name of each ssd hard disks in group, and according to the cryptographic Hash of the filename of each focus file and each
The cryptographic Hash and preset rules of the hard disk name of the ssd hard disks, each focus file is dumped into corresponding institute respectively
State in ssd hard disks.Specifically, referring to Fig. 4, S305 implementation procedure is as follows：
S401：According to uniformity hash algorithm, to each memory space less than the hard disk names of the ssd hard disks breathe out
It is uncommon, calculate each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each memory space less than
The cryptographic Hash of the hard disk name of the ssd hard disks, each memory space less than the ssd hard disks form a Hash ring, obtain
Each memory space less than position of the ssd hard disks on the Hash ring；
S402：According to uniformity hash algorithm, Hash is carried out to the filename of each focus file, calculated each
The cryptographic Hash of the filename of the focus file；
S403：, will be each according to the filename of each focus file and the cryptographic Hash of the hard disk name of each ssd hard disks
The focus file is dumped in the corresponding ssd hard disks on the Hash ring, makes the Kazakhstan of the filename of each focus file
The difference of uncommon value and the cryptographic Hash of the hard disk name of respective stored ssd hard disks is minimum, and the filename of each focus file
Cryptographic Hash is less than the cryptographic Hash of the hard disk name of corresponding storage ssd hard disks.
Specifically, S305 specific implementation principle is identical with the implementation principle of S103 in above-described embodiment.S305 correlation
Explanation refers to the above-mentioned explanation to S103, will not be repeated here.
, will when the hard disk for the cryptographic Hash that the filename more than focus file is not present in the Hash ring in ssd hard disk groups
The focus file is stored on the starting hard disk in ssd hard disk groups Hash rings.
The storage method for the CDN fringe node files that the present embodiment provides, determines the multiple heat stored in sata hard disk groups
Dot file, and calculated according to uniformity hash algorithm in each cryptographic Hash and ssd hard disk groups of the filename of the focus file
The cryptographic Hash of the hard disk name of each ssd hard disks, and cryptographic Hash according to the filename of each focus file and it is each described in
The cryptographic Hash and preset rules of the hard disk name of ssd hard disks, each focus file is dumped into the corresponding ssd respectively
In hard disk, the pressure of sata hard disks is reduced, improves the utilization rate of ssd hard disks, each ssd disks are individually present and storage file, i.e.,
When making the faulty disk of two pieces and more than two pieces of appearance, will not also cause whole server unavailable, and changing faulty disk, weight is only needed
New synchronous edge node server belongs to the data of the faulty disk, and fault time is greatly shortened, improves the stabilization of system
Property.
Based on a kind of storage method of CDN fringe nodes file disclosed in above-described embodiment, referring to Fig. 5, the present embodiment
It is corresponding to disclose a kind of storage device of CDN fringe nodes file, including：
Acquiring unit 101, the raid card of Nonraid patterns is supported for obtaining, and makes hard disk groups by the raid card
It is connected with edge node server；
Initialization unit 102, for being initialized to the hard disk groups, obtain the hard of each hard disk in the hard disk groups
Disk name；
Specifically, the initialization unit 102 includes：
Subelement is identified, for each hard disk in the hard disk groups to be identified, determines each hard disk respectively
Equipment symbol title and type, the hard disk type be sata hard disks or ssd hard disks；
Subregion subelement is formatted, for being formatted subregion to each hard disk respectively；
Subelement is established, for the quantity according to hard disk in the hard disk groups, establishes hard disk serial；
Carry subelement, for according to the traversal order to each hard disk, being respectively mounted to each hard disk
In the hard disk serial, and the hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
First memory cell 103, for calculating the hard disk of each hard disk in the hard disk groups according to uniformity hash algorithm
The cryptographic Hash of the filename of name and each file to be stored, and according to the hard disk name of each hard disk and each described to be stored
The cryptographic Hash and preset rules of the filename of file, the corresponding hard disk is arrived into each file to be stored storage respectively
In；
Specifically, first memory cell 103 includes：
First computation subunit, for according to uniformity hash algorithm, to each memory space less than the hard disk
Hard disk name carry out Hash, calculate each memory space less than the hard disk hard disk name cryptographic Hash, according to each storage
Space less than the hard disk hard disk name cryptographic Hash, each memory space less than the hard disk form a Hash ring,
Obtain each memory space less than position of the hard disk on the Hash ring；
Second computation subunit, for according to uniformity hash algorithm, entering to the filename of each file to be stored
Row Hash, calculate the cryptographic Hash of the filename of each file to be stored；
First storing sub-units, the hard disk for the filename according to each file to be stored and each hard disk
The cryptographic Hash of name, by the storage of each file to be stored into the corresponding hard disk on the Hash ring, make each text to be stored
The difference of the cryptographic Hash of the filename of part and the cryptographic Hash of the hard disk name of respective stored hard disk is minimum, and each text to be stored
The cryptographic Hash of the filename of part is less than the cryptographic Hash of the hard disk name of corresponding storage hard disk.
The present embodiment provide CDN fringe node files storage device, from support Nonraid patterns raid card,
Every piece of hard disk is in the state of independent operating, and avoiding establishment raid5 in the prior art needs one piece of hard disk to make redundancy disk, improves
The maximum capacity of edge node server, covers more files, reduces back source request, improves Consumer's Experience；This reality
Hard disk name and the cryptographic Hash and preset rules of the filename of each file to be stored of the example according to each hard disk are applied,
Each file to be stored is stored into the corresponding hard disk respectively, it is each to ensure that file to be stored is uniformly stored in
In hard disk, because the present embodiment does not need one piece of hard disk to make redundancy disk, each hard disk is individually present and storage file, even if occurring
When more than two pieces of faulty disk, will not also cause whole server unavailable, and change faulty disk, the re-synchronization edge is only needed
Node server belongs to the data of the faulty disk, and fault time is greatly shortened, and improves the stability of system.
Referring to Fig. 6, present embodiment discloses the storage device of another CDN fringe nodes file, including：
Acquiring unit 201, the raid card of Nonraid patterns is supported for obtaining, and makes sata hard by the raid card
Disk group and ssd hard disk groups are connected with edge node server；
Initialization unit 202, for being initialized to the sata hard disk groups and the ssd hard disk groups, obtain described
The hard disk name of each ssd hard disks in hard disk groups；
First memory cell 203, for calculating each sata hard disks in the hard disk groups according to uniformity hash algorithm
The cryptographic Hash of the filename of hard disk name and each file to be stored, and the hard disk name according to each sata hard disks and each institute
The cryptographic Hash and preset rules of the filename of file to be stored are stated, respectively arrives each file to be stored storage accordingly
In the sata hard disks；
Determining unit 204, in preset time period, it is determined that in the file stored in the sata hard disk groups
Multiple focus files；
Specifically, the determining unit 204 includes：
Subelement is counted, in preset time period, counting the visit of each file stored in the sata hard disk groups
Ask the frequency；
Sort subelement, for being carried out from high to low to each file stored in the sata hard disk groups by visitation frequency
Sequence；
Determination subelement, for the file of the high visitation frequency of preset ratio in the sequence to be defined as into focus file.
Second memory cell 205, for the filename according to each focus file of uniformity hash algorithm calculating
The cryptographic Hash of the hard disk name of each ssd hard disks in cryptographic Hash and ssd hard disk groups, and according to the filename of each focus file
Cryptographic Hash and each ssd hard disks hard disk name cryptographic Hash and preset rules, respectively will each focus file
Dump in the corresponding ssd hard disks.
Specifically, second memory cell 205 includes：
3rd computation subunit, for according to uniformity hash algorithm, to each memory space less than the ssd hard disks
Hard disk name carry out Hash, calculate each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each
Memory space less than the ssd hard disks hard disk name cryptographic Hash, each memory space less than the ssd hard disks form one
Individual Hash ring, obtain each memory space less than position of the ssd hard disks on the Hash ring；
4th computation subunit, for according to uniformity hash algorithm, being carried out to the filename of each focus file
Hash, calculate the cryptographic Hash of the filename of each focus file；
Second storing sub-units, the hard disk name for filename and each ssd hard disks according to each focus file
Cryptographic Hash, each focus file is dumped in the corresponding ssd hard disks on the Hash ring, makes each focus text
The difference of the cryptographic Hash of the filename of part and the cryptographic Hash of the hard disk name of respective stored ssd hard disks is minimum, and each focus
The cryptographic Hash of the filename of file is less than the cryptographic Hash of the hard disk name of corresponding storage ssd hard disks.
It should be noted that the storage device of the CDN fringe nodes file also includes：
3rd memory cell, for when the Hash that the filename more than file to be stored is not present in corresponding Hash ring
During the hard disk of the cryptographic Hash of the filename of value or focus file, the file to be stored or the focus file are stored in accordingly
On starting hard disk in Hash ring.
The storage device for the CDN fringe node files that the present embodiment provides, determines the multiple heat stored in sata hard disk groups
Dot file, and according to the Hash of the cryptographic Hash of the filename of each focus file and the hard disk name of each ssd hard disks
Value and preset rules, each focus file is dumped in the corresponding ssd hard disks respectively, reduces sata hard disks
Pressure, the utilization rate of ssd hard disks is improved, each ssd disks are individually present and storage file, even if occurring two pieces and more than two pieces
When faulty disk, will not also cause whole server unavailable, and change faulty disk, the re-synchronization edge node server is only needed
Belong to the data of the faulty disk, fault time is greatly shortened, improve the stability of system.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.For device disclosed in embodiment
For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part
It is bright.
It should also be noted that, herein, such as first and second or the like relational terms are used merely to one
Entity or operation make a distinction with another entity or operation, and not necessarily require or imply between these entities or operation
Any this actual relation or order be present.Moreover, term " comprising ", "comprising" or its any other variant are intended to contain
Lid nonexcludability includes, so that process, method, article or equipment including a series of elements not only will including those
Element, but also the other element including being not expressly set out, or it is this process, method, article or equipment also to include
Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that
Other identical element also be present in process, method, article or equipment including the key element.
Directly it can be held with reference to the step of method or algorithm that the embodiments described herein describes with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), internal memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (14)
- A kind of 1. storage method of CDN fringe nodes file, it is characterised in that including：The raid card for supporting Nonraid patterns is obtained, and hard disk groups is connected with edge node server by the raid card；The hard disk groups are initialized, obtain the hard disk name of each hard disk in the hard disk groups；Each hard disk name of hard disk and the file of each file to be stored in the hard disk groups are calculated according to uniformity hash algorithm The cryptographic Hash of name, and according to the cryptographic Hash of the hard disk name of each hard disk and the filename of each file to be stored and Preset rules, each file to be stored is stored into the corresponding hard disk respectively.
- 2. according to the method for claim 1, it is characterised in that it is described that the hard disk groups are initialized, obtain described The hard disk name of each hard disk in hard disk groups, including：Each hard disk in the hard disk groups is identified, determines the equipment symbol title and type of each hard disk respectively, The hard disk type is sata hard disks or ssd hard disks；Subregion is formatted to each hard disk respectively；According to the quantity of hard disk in the hard disk groups, hard disk serial is established；According to the traversal order to each hard disk, each hard disk is mounted in the hard disk serial respectively, And the hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
- 3. according to the method for claim 1, it is characterised in that described to calculate the hard disk groups according to uniformity hash algorithm In each hard disk hard disk name and each file to be stored filename cryptographic Hash, and according to the hard disk name of each hard disk , respectively will each file to be stored storage with the cryptographic Hash and preset rules of the filename of each file to be stored Into the corresponding hard disk, including：According to uniformity hash algorithm, to each memory space less than the hard disk name of the hard disk carry out Hash, calculate every Individual memory space less than the hard disk hard disk name cryptographic Hash, according to each memory space less than the hard disk hard disk Name cryptographic Hash, each memory space less than the hard disk form a Hash ring, obtain each memory space less than institute State position of the hard disk on the Hash ring；According to uniformity hash algorithm, Hash is carried out to the filename of each file to be stored, calculates each described treat The cryptographic Hash of the filename of storage file；, will be each to be stored according to the filename of each file to be stored and the cryptographic Hash of the hard disk name of each hard disk File is stored into the corresponding hard disk on the Hash ring, make each file to be stored filename cryptographic Hash with it is corresponding The difference of the cryptographic Hash of the hard disk name of storage hard disk is minimum, and the cryptographic Hash of the filename of each file to be stored is less than pair Answer the cryptographic Hash of the hard disk name of storage hard disk.
- 4. according to the method for claim 1, it is characterised in that the hard disk groups are sata hard disk groups, respectively will described After each file to be stored storage is into the corresponding hard disk, methods described also includes：In preset time period, it is determined that multiple focus files in the file stored in sata hard disk groups；Calculated according to uniformity hash algorithm each in the cryptographic Hash and ssd hard disk groups of the filename of each focus file The cryptographic Hash of the hard disk name of ssd hard disks, and it is hard according to the cryptographic Hash and each ssd of the filename of each focus file The cryptographic Hash and preset rules of the hard disk name of disk, each focus file is dumped into the corresponding ssd hard disks respectively In.
- 5. according to the method for claim 4, it is characterised in that it is described in preset time period, it is determined that in sata hard disk groups In multiple focus files in the file that has stored, including：In preset time period, the visitation frequency of each file stored in the sata hard disk groups is counted；The each file stored in the sata hard disk groups is ranked up from high to low by visitation frequency；The file of the high visitation frequency of preset ratio in the sequence is defined as focus file.
- 6. according to the method for claim 4, it is characterised in that described to calculate each heat according to uniformity hash algorithm The cryptographic Hash of the hard disk name of each ssd hard disks in the cryptographic Hash and ssd hard disk groups of the filename of dot file, and according to each described The cryptographic Hash and preset rules of the cryptographic Hash of the filename of focus file and the hard disk name of each ssd hard disks, respectively will Each focus file is dumped in the corresponding ssd hard disks, including：According to uniformity hash algorithm, to each memory space less than the hard disk names of the ssd hard disks carry out Hash, calculate Each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each memory space less than the ssd it is hard The cryptographic Hash of the hard disk name of disk, each memory space less than the ssd hard disks form a Hash ring, it is empty to obtain each storage Between less than position of the ssd hard disks on the Hash ring；According to uniformity hash algorithm, Hash is carried out to the filename of each focus file, calculates each focus The cryptographic Hash of the filename of file；, will each focus according to the filename of each focus file and the cryptographic Hash of the hard disk name of each ssd hard disks File is dumped in the corresponding ssd hard disks on the Hash ring, makes the cryptographic Hash and phase of the filename of each focus file The difference minimum of the cryptographic Hash of the hard disk name of ssd hard disks should be stored, and the cryptographic Hash of the filename of each focus file is small In the cryptographic Hash of the hard disk name of corresponding storage ssd hard disks.
- 7. the method according to claim 3 or 6, it is characterised in that methods described also includes：When in corresponding Hash ring be not present more than file to be stored the cryptographic Hash of filename or the filename of focus file Cryptographic Hash hard disk when, the starting hard disk that the file to be stored or the focus file is stored in corresponding Hash ring On.
- A kind of 8. storage device of CDN fringe nodes file, it is characterised in that including：Acquiring unit, the raid card of Nonraid patterns is supported for obtaining, and makes hard disk groups and edge section by the raid card Point server is connected；Initialization unit, for being initialized to the hard disk groups, obtain the hard disk name of each hard disk in the hard disk groups；First memory cell, for the hard disk name according to each hard disk in the uniformity hash algorithm calculating hard disk groups and each The cryptographic Hash of the filename of file to be stored, and according to the hard disk name of each hard disk and the text of each file to be stored The cryptographic Hash and preset rules of part name, each file to be stored is stored into the corresponding hard disk respectively.
- 9. device according to claim 8, it is characterised in that the initialization unit includes：Subelement is identified, for each hard disk in the hard disk groups to be identified, determines setting for each hard disk respectively Standby symbol title and type, the hard disk type are sata hard disks or ssd hard disks；Subregion subelement is formatted, for being formatted subregion to each hard disk respectively；Subelement is established, for the quantity according to hard disk in the hard disk groups, establishes hard disk serial；Carry subelement, for according to the traversal order to each hard disk, being respectively mounted to each hard disk described In hard disk serial, and the hard disk name using the sequential value of each hard disk as corresponding hard disk respectively.
- 10. device according to claim 8, it is characterised in that first memory cell includes：First computation subunit, for according to uniformity hash algorithm, to each memory space less than the hard disk hard disk Name carry out Hash, calculate each memory space less than the hard disk hard disk name cryptographic Hash, according to each memory space Less than the hard disk hard disk name cryptographic Hash, each memory space less than the hard disk form a Hash ring, obtain Each memory space less than position of the hard disk on the Hash ring；Second computation subunit, for according to uniformity hash algorithm, being breathed out to the filename of each file to be stored It is uncommon, calculate the cryptographic Hash of the filename of each file to be stored；First storing sub-units, for the filename according to each file to be stored and the hard disk name of each hard disk Cryptographic Hash, by the storage of each file to be stored into the corresponding hard disk on the Hash ring, make each file to be stored The difference of the cryptographic Hash of filename and the cryptographic Hash of the hard disk name of respective stored hard disk is minimum, and each file to be stored The cryptographic Hash of filename is less than the cryptographic Hash of the hard disk name of corresponding storage hard disk.
- 11. device according to claim 8, it is characterised in that the hard disk groups are sata hard disk groups, and described device is also wrapped Include：Determining unit, in preset time period, it is determined that multiple focuses text in the file stored in sata hard disk groups Part；Second memory cell, for filename that each focus file is calculated according to uniformity hash algorithm cryptographic Hash and The cryptographic Hash of the hard disk name of each ssd hard disks in ssd hard disk groups, and the cryptographic Hash of the filename according to each focus file With the cryptographic Hash and preset rules of the hard disk name of each ssd hard disks, each focus file is dumped into phase respectively In the ssd hard disks answered.
- 12. device according to claim 11, it is characterised in that the determining unit includes：Subelement is counted, in preset time period, counting the access frequency of each file stored in the sata hard disk groups It is secondary；Sort subelement, for being arranged from high to low each file stored in the sata hard disk groups by visitation frequency Sequence；Determination subelement, for the file of the high visitation frequency of preset ratio in the sequence to be defined as into focus file.
- 13. device according to claim 11, it is characterised in that second memory cell includes：3rd computation subunit, for according to uniformity hash algorithm, to each memory space less than the ssd hard disks it is hard Disk name carry out Hash, calculate each memory space less than the ssd hard disks hard disk name cryptographic Hash, according to each storage Space less than the ssd hard disks hard disk name cryptographic Hash, each memory space less than the ssd hard disks form one Kazakhstan Uncommon ring, obtain each memory space less than position of the ssd hard disks on the Hash ring；4th computation subunit, for foundation uniformity hash algorithm, Hash is carried out to the filename of each focus file, Calculate the cryptographic Hash of the filename of each focus file；Second storing sub-units, for the filename according to each focus file and the Kazakhstan of the hard disk name of each ssd hard disks Uncommon value, each focus file is dumped in the corresponding ssd hard disks on the Hash ring, makes each focus file The difference of the cryptographic Hash of filename and the cryptographic Hash of the hard disk name of respective stored ssd hard disks is minimum, and each focus file Filename cryptographic Hash be less than corresponding storage ssd hard disks hard disk name cryptographic Hash.
- 14. the device according to claim 10 or 13, it is characterised in that described device also includes：3rd memory cell, for when in corresponding Hash ring be not present more than file to be stored filename cryptographic Hash or During the hard disk of the cryptographic Hash of the filename of focus file, the file to be stored or the focus file are stored in corresponding Hash On starting hard disk in ring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710752436.XA CN107463342B (en) | 2017-08-28 | 2017-08-28 | CDN edge node file storage method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710752436.XA CN107463342B (en) | 2017-08-28 | 2017-08-28 | CDN edge node file storage method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107463342A true CN107463342A (en) | 2017-12-12 |
| CN107463342B CN107463342B (en) | 2021-04-20 |
Family
ID=60550642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710752436.XA Active CN107463342B (en) | 2017-08-28 | 2017-08-28 | CDN edge node file storage method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107463342B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109086007A (en) * | 2018-07-25 | 2018-12-25 | 郑州云海信息技术有限公司 | The method and system that a kind of pair of polymorphic type hard disk is formatted |
| CN109525662A (en) * | 2018-11-14 | 2019-03-26 | 程桂平 | The method of copy is set for Hot Contents |
| CN109683816A (en) * | 2018-12-14 | 2019-04-26 | 北京奇艺世纪科技有限公司 | The disk configuration method and system of a kind of time source tree node |
| CN110502480A (en) * | 2019-08-16 | 2019-11-26 | 济南浪潮数据技术有限公司 | Distributed storage data retrieval method, system, device and readable storage medium storing program for executing |
| CN111770152A (en) * | 2020-06-24 | 2020-10-13 | 中国科学院计算技术研究所 | Edge data management method, medium, edge server and system |
| CN111930316A (en) * | 2020-09-09 | 2020-11-13 | 上海七牛信息技术有限公司 | Cache read-write system and method for content distribution network |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2498476A2 (en) * | 2011-03-08 | 2012-09-12 | Rackspace Us, Inc. | Massively scalable object storage system |
| CN103761167A (en) * | 2014-01-23 | 2014-04-30 | 浪潮(北京)电子信息产业有限公司 | Method and device for achieving data center backup |
| CN104202407A (en) * | 2014-09-10 | 2014-12-10 | 北京奇艺世纪科技有限公司 | Video file synchronization method and video file synchronization device |
| CN104469392A (en) * | 2014-12-19 | 2015-03-25 | 北京奇艺世纪科技有限公司 | Video file storage method and device |
| CN104935648A (en) * | 2015-06-03 | 2015-09-23 | 北京快网科技有限公司 | High-cost-performance CDN system, and file pre-push and fragment buffer memory methods |
| CN105262841A (en) * | 2015-11-06 | 2016-01-20 | 浪潮软件集团有限公司 | CDN network load balancing implementation method and CDN scheduling server |
| US20160352730A1 (en) * | 2015-05-28 | 2016-12-01 | Pearson Education, Inc. | Data access and anonymity management |
| CN106790324A (en) * | 2015-11-20 | 2017-05-31 | 华为技术有限公司 | Content distribution method, virtual server management method, cloud platform and system |
| CN106990915A (en) * | 2017-02-27 | 2017-07-28 | 北京航空航天大学 | A kind of SRM method based on storage media types and weighting quota |
-
2017
- 2017-08-28 CN CN201710752436.XA patent/CN107463342B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2498476A2 (en) * | 2011-03-08 | 2012-09-12 | Rackspace Us, Inc. | Massively scalable object storage system |
| CN103761167A (en) * | 2014-01-23 | 2014-04-30 | 浪潮(北京)电子信息产业有限公司 | Method and device for achieving data center backup |
| CN104202407A (en) * | 2014-09-10 | 2014-12-10 | 北京奇艺世纪科技有限公司 | Video file synchronization method and video file synchronization device |
| CN104469392A (en) * | 2014-12-19 | 2015-03-25 | 北京奇艺世纪科技有限公司 | Video file storage method and device |
| US20160352730A1 (en) * | 2015-05-28 | 2016-12-01 | Pearson Education, Inc. | Data access and anonymity management |
| CN104935648A (en) * | 2015-06-03 | 2015-09-23 | 北京快网科技有限公司 | High-cost-performance CDN system, and file pre-push and fragment buffer memory methods |
| CN105262841A (en) * | 2015-11-06 | 2016-01-20 | 浪潮软件集团有限公司 | CDN network load balancing implementation method and CDN scheduling server |
| CN106790324A (en) * | 2015-11-20 | 2017-05-31 | 华为技术有限公司 | Content distribution method, virtual server management method, cloud platform and system |
| CN106990915A (en) * | 2017-02-27 | 2017-07-28 | 北京航空航天大学 | A kind of SRM method based on storage media types and weighting quota |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109086007A (en) * | 2018-07-25 | 2018-12-25 | 郑州云海信息技术有限公司 | The method and system that a kind of pair of polymorphic type hard disk is formatted |
| CN109525662A (en) * | 2018-11-14 | 2019-03-26 | 程桂平 | The method of copy is set for Hot Contents |
| CN109683816A (en) * | 2018-12-14 | 2019-04-26 | 北京奇艺世纪科技有限公司 | The disk configuration method and system of a kind of time source tree node |
| CN109683816B (en) * | 2018-12-14 | 2021-08-27 | 北京奇艺世纪科技有限公司 | Disk configuration method and system for back source tree nodes |
| CN110502480A (en) * | 2019-08-16 | 2019-11-26 | 济南浪潮数据技术有限公司 | Distributed storage data retrieval method, system, device and readable storage medium storing program for executing |
| CN110502480B (en) * | 2019-08-16 | 2022-04-29 | 济南浪潮数据技术有限公司 | Distributed storage data retrieval method, system, device and readable storage medium |
| CN111770152A (en) * | 2020-06-24 | 2020-10-13 | 中国科学院计算技术研究所 | Edge data management method, medium, edge server and system |
| CN111770152B (en) * | 2020-06-24 | 2021-07-20 | 中国科学院计算技术研究所 | Edge data management method, medium, edge server and system |
| CN111930316A (en) * | 2020-09-09 | 2020-11-13 | 上海七牛信息技术有限公司 | Cache read-write system and method for content distribution network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107463342B (en) | 2021-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107463342A (en) | A kind of storage method and device of CDN fringe nodes file | |
| US10776396B2 (en) | Computer implemented method for dynamic sharding | |
| CN103763383B (en) | Integrated cloud storage system and its storage method | |
| US20160226931A9 (en) | Virtual multi-cluster clouds | |
| CN102855294A (en) | Intelligent hash data layout method, cluster storage system and method thereof | |
| CN101187931A (en) | Distribution type file system multi-file copy management method | |
| CN104281506A (en) | Data maintenance method and system for file system | |
| CN105025053A (en) | Distributed file upload method based on cloud storage technology and system | |
| CN104008152A (en) | Distributed file system architectural method supporting mass data access | |
| CN103929500A (en) | Method for data fragmentation of distributed storage system | |
| CN103929454A (en) | Load balancing storage method and system in cloud computing platform | |
| CN103535014B (en) | A kind of network store system, data processing method and client | |
| CN103067525A (en) | Cloud storage data backup method based on characteristic codes | |
| CN108512908A (en) | A kind of cloud storage fault tolerant mechanism based on Ceph and the web-based management platform based on Ceph | |
| CN104462389A (en) | Method for implementing distributed file systems on basis of hierarchical storage | |
| CN103631666A (en) | Data redundancy fault-tolerance adaptation management device, service device, system and method | |
| CN104954444B (en) | A kind of method and apparatus that migration is data cached | |
| CN102546823A (en) | File storage management system of cloud storage system | |
| CN106445409A (en) | Distributed block storage data writing method and device | |
| CN102982182A (en) | Data storage planning method and device | |
| CN110188084A (en) | A kind of distributed file storage system and file memory method | |
| CN107340975A (en) | A kind of method and device of file storage | |
| CN105827678A (en) | High-availability framework based communication method and node | |
| CN111026711A (en) | Block chain based data storage method and device, computer equipment and storage medium | |
| CN109117292A (en) | A kind of method and apparatus and cluster storage system of cluster-based storage |
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 |