CN104391901B - A kind of memory cell network big data base frame platform and its file access method - Google Patents
A kind of memory cell network big data base frame platform and its file access method Download PDFInfo
- Publication number
- CN104391901B CN104391901B CN201410644353.5A CN201410644353A CN104391901B CN 104391901 B CN104391901 B CN 104391901B CN 201410644353 A CN201410644353 A CN 201410644353A CN 104391901 B CN104391901 B CN 104391901B
- Authority
- CN
- China
- Prior art keywords
- file
- user
- server
- hash values
- node
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2219—Large Object storage; Management thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of memory cell network big data base frame platform and its file access method, the storage managing server of the multiple operators connected including internet and the terminal user for being registered in different operators, storage managing server to connect and be connected to HASH values server and user's stored record server on multiple node store servers, internet；When user sets up or updates network file, the network file is produced by HASH values server first and a current time HASH value is assigned, file sitation index is set up in user's stored record server to the file for assigning current time HASH values simultaneously, the current time HASH values are unique newest HASH values, and user is according to newest HASH values extraction document.The present invention forms a kind of new network file access structures and methods, has abandoned traditional direct-connected storage, can save traditional file backup between each operator, save the bandwidth usage amount of network.
Description
Technical field
The present invention relates to data network transmission, more particularly to a kind of memory cell network big data base frame platform and
Its file access method is a kind of by the network big data base frame platform of ARM memory cell and its file access side
Method.
Background technology
ARM frameworks, the past is referred to as Advanced Reduced Instruction Set machine（Advanced RISC Machine, are referred to as earlier：
Acorn RISC Machine）, it is 32 reduced instruction set computers（RISC）Processor architecture, it is widely used in many embedding
Embedded system is designed.The characteristics of due to energy-conservation, arm processor is highly suitable for moving communicating field, meets its major design mesh
It is designated as low cost, high-performance, the characteristic of low power consumption.
ARM platforms are generally used for information electronic device, special system and industrial control field, such as：Panel computer, mobile phone,
Automobile computer, automatically control, as the CPU disposal abilities under ARM frameworks constantly strengthen, function is gradually become strong, and interface is continuous
Increase, some companies research and develop server product by the CPU of ARM platforms, pass through the processing of tens to hundreds of CPU simultaneously
Ability carries out computing, because the CPU energy consumptions under ARM frameworks are very low, so by the server of tens to hundreds of CPU composition,
Performance is as the server under traditional X86-based, and power consumption consumption is upper advantageous.
The direct-connected storage (Direct-Attached Storage, abbreviation DAS) of traditional storing open system is
There is the usage history of nearly 40 years and continue up and be applied to big data base frame platform so far, it is continuous with user data
Increase, during especially hundreds of more than GB, it becomes increasingly to perplex system pipes the problem of in terms of backup, recovery, extension, calamity are standby
Reason person.
Interface channel between direct-connected storage and server host generally uses SCSI connections, with server CPU's
Disposal ability is more and more stronger, and storage hard disk space is increasing, and the hard disk quantity of array is more and more, and SCSI passages will turn into
I/O bottleneck；Server host SCSI ID resource-constraineds, the SCSI passages connection that can be set up is limited.But also there was only this
Proprietary function, it is impossible to be then turned on other functions, wastes power consumption, hardware resource and cost of investment, is not that today's society is carried
Advocate the theory of energy-conserving and environment-protective.
The content of the invention
A kind of memory cell network big data base that the problem of purpose of the present invention is and is directed to above-mentioned direct-connected storage proposes
Plinth framework platform and its file access method.
To achieve these goals, the technical scheme is that：
It is connected on a kind of memory cell network big data base frame platform, including internet, internet multiple
The database of operator and the terminal user for being registered in different operators, database include multiple node store servers, node
Storage server connects data storage cell；Wherein, HASH values server and user are also associated with respectively on the internet
Stored record server；The HASH values server is used to upload terminal user or more new file assigns a HASH and is newly worth simultaneously
Storage, user's stored record server storage assigns the searching route that HASH is newly worth file.
Scheme is further：The node store server is the node store server of ARM frameworks, including ARM frameworks
Central processing unit；The central processing unit includes PCIE series interfaces circuit and SATA series interfaces circuit and RGM series
Interface circuit；The PCIE series interfaces circuit turns SATA bridge by a PCIE and connects circuit at least two SATA hard disc numbers of connection
According to memory cell, the SATA series interfaces circuit connects at least two SATA hard disc data by a SATA expanded circuit and deposited
Storage unit, the RGM series interfaces circuit is connected through network PHY with internet.
Scheme is further：The database also includes storage managing server, the storage of storage managing server connecting node
Server is deposited, the storage managing server is used for the hard disk service condition of monitor node, dispatches the usage amount of hard-disc storage, and
Balanced fixed disk file storage.
A kind of network file access method, is the file access side based on memory cell network big data base frame platform
Method, the platform includes being connected to the storage managing server of multiple operators on internet, internet and is registered in not
With the terminal user of operator, storage managing server connects multiple node store servers, node store server connections
According to memory cell；HASH values server and user's stored record server are also associated with the internet respectively；Wherein, institute
Stating network file access method is：When user sets up or updates network file, first by HASH values server to the net
Network file produces and assigns a current time HASH value, remembers while being stored to the file for assigning current time HASH values in user
File sitation index is set up in record server, the current time HASH values are unique newest HASH values, and user is according to newest
HASH value extraction documents.
Scheme is further：The user is according to the process of newest HASH values extraction document：When terminal user sends text
After part processing request, by the storage managing server under current operator where the request analysis of terminal user to user, use
The catalogue file list of oneself is set up or opened at family after registering, logging in, and finds the file that oneself download or update, and clicking on will
The file of download, is found by this document HASH values stored in user's stored record server for being connected with internet first
Corresponding link, the HASH values of this document if on the node memory where this document link is the operator that user logs in
It is newest, then this document will directly download；If on the node for the operator that this document is not logged in user,
By link by the node memory of the file download of user to operator where user come then user downloads again.
Scheme is further：It is described file sitation index is set up in user's stored record server process be：When with
Family is set up after a new file or more new file, and obtaining one by the HASH values server being arranged on internet first works as
The HASH values of preceding time assign new file or more new file, by the current operation of imparting current time HASH values file deposit
On node memory where business, the chain of node is grounded where assigning current operator to the current time HASH values file
Location；Filename, current time HASH values and chained address are stored in the user's stored record server life being connected with internet
Into sitation index, and when filename is identical, HASH values and the chained address of same file name are updated.
Scheme is further：The node store server is the node store server of ARM frameworks, including ARM frameworks
Central processing unit；The central processing unit includes PCI-E series interfaces circuit and SATA series interfaces circuit and RGM series
Interface circuit；The PCI-E series interfaces circuit turns SATA bridge by a PCI-E and connects circuit at least two SATA hard discs of connection
Data storage cell, the SATA series interfaces circuit connects at least two SATA hard disc data by a SATA expanded circuit
Memory cell, the RGM series interfaces circuit is connected through network PHY with internet.
Scheme is further：The HASH values server and the family stored record server are by the way of two-node cluster hot backup
Work.
Scheme is further：The network file access method, which further comprises accessing node store server file, supervises
Control；The method of monitoring is：The networking state of storage managing server real-time detection node store server, monitoring node storage clothes
The shape that cpu load, memory size and utilization rate, memory capacity and utilization rate, the network bandwidth and the utilization rate, hard disk of device of being engaged in are run
Condition information；
When the network traffics and storage usage amount of some node are more than or equal to the 80% of usage amount, the access of file is adjusted
Spend under other nodes of operator；
When a node does not have handshaking information or delay of response, the access of file is dispatched to other nodes of operator
Under, until node recovers.
Scheme is further：The operator includes by current operator of the request end source address based on and ask the visitor in
End source address is sought by other operators for exporting.
Compared with the prior art, the invention has the advantages that：
1, HASH（Hash）The setting of value server and user's stored record server forms a kind of new network file
Access structure and method, have abandoned traditional direct-connected storage (Direct-Attached Storage, abbreviation DAS), each
Individual file is identified using unique HASH values, and the introducing with HASH values and chained address index, make whenever, it is any
The file that point is opened all is newest, and the memory of any operator can serve as the storage of this document, can save
The traditional file backup of each operator, new HASH values are set and the bar of retrieval mode reduces file with file in online weight
Multiple transmission, and the congestion come to Netowrk tape is corrected, improve the saturation of network.
The introducing of 2.ARM architectural node storage servers, solves and utilizes different pieces of information passage high-speed read-write hard disc data
The problem of；The resource of CPU, internal memory, hard disk, network interface is taken full advantage of, them has been reached unified quick call capability,
One ARM architectural nodes storage server can connect 4 4T capacity HDDs simultaneously, play maximum efficiency；In equivalent capability
Under, cost is lower by 70% than X86-based node store server, low in energy consumption 80%, small volume 50%.
The present invention is described in detail with reference to the accompanying drawings and examples.
Brief description of the drawings
Fig. 1 is inventive network platform topology figure；
Fig. 2 is node store server architecture schematic diagram of the present invention；
Fig. 3 is file access method workflow diagram of the present invention.
Embodiment
Embodiment 1：
A kind of memory cell network big data base frame platform, as shown in figure 1, including internet 1, on internet
It is connected to multiple operator A, B, C（Such as operator of UNICOM A, telecom operators B, mobile operator C）Database and be registered in
The terminal user 2 of different operators, database includes multiple node store servers 3, and node store server connection data is deposited
Storage unit 4；Wherein, HASH is also associated with respectively on the internet（Hash）It is worth server 5 and the service of user's stored record
Device 6；The HASH values server is used to upload terminal user or more new file assigns a HASH and is newly worth and stores, described to use
Family stored record server storage assigns the searching route that HASH is newly worth file.
In order to increase memory capacity, described node store server utility has multiple, in order to be taken to multiple node stores
Business device is managed, and the database also includes storage managing server 7, and storage managing server is stored up by network connection node
Server is deposited, the storage managing server is used for the hard disk service condition of monitor node, dispatches the usage amount of hard-disc storage, and
Balanced fixed disk file storage.
In embodiment：As shown in Fig. 2 the node store server is the node store server of ARM frameworks, including
ARM frameworks central processing unit 301；The central processing unit includes PCI-E series interfaces circuit and SATA series interfaces circuits
And RGM series interfaces circuits；The PCI-E series interfaces circuit turns SATA bridge by a PCI-E and connects circuit 302 to be connected to
Few two data storage cells of SATA hard disc 303, the SATA series interfaces circuit is connected by a SATA expanded circuit 304
The data storage cell of at least two SATA hard disc 303, the RGM series interfaces circuit is connected through network PHY with internet.
Central processing unit in embodiment is ARM Cortex-A9 four core processors, and CPU highest frequency reaches
1.0GHz, supports the bus bandwidth of 64,1066MHz DDR3 data bus interfaces, throughput rate faster, maximum memory capacity
Support 4GB.
Realize multiplex SATA interface by the signals of USB 2.0 to transfer on usual ARM platforms.But this can only be solved
Certainly interface problem, can not meet the read or write speed of user data in terms of speed.In order to meet the memory space of Large Copacity, and energy
Enough meet high-speed data read or write speed, it is necessary to need using the interfaces of standard SATA II 2.0 conventional at present.Embodiment selects 1
The CPU of the 1x interfaces of road PCI-E 2.0 and 1 road SATA II interfaces, while the fastest of PCI-E 2.0 can reach 6Gb/s,
SATA II speed can also reach 6Gb/s.
At present, the I/O read or write speeds of conventional enterprise-level hard disk are most fast in 160MB/s, and standard SATA II signals
Read or write speed is at least in more than 350MB/s, and the transmission speed of hard disk is maximum bottleneck, in order to eliminate or reduce the bottle of hard disk
Neck, the SATA II interfaces of 2 tunnel standards are extended out using PCI-E 2.0 bridging chip, while connecing 6Gb/s SATA II
Mouth connects hard disk all the way, and such a ARM CPU can be with external 3 hard disks.Massive store is realized, also meets and passes at a high speed
The requirement of transmission of data, while also solving the bottleneck of hard disk transmission speed.
The present embodiment CPU connects the SATA II interfaces that bridging chip extends out two standards by PCI-E-2.0 signals, expands
The big memory space of equipment, while can also meet user quickly reads related data.CPU passes through SATA-II signals simultaneously
Connection extended chip extends out the SATA-II interfaces of 2 standards.The CPU device of the single ARM platforms of this sample embodiment just can be simultaneously
Connect 4 SATA hard discs.
In embodiment bridging chip be first PCI-E signal after PCI-E PHY processing some useless signals
Separate, while analyzing useful data in the packet of transmission, these useful data are passed through with useful signaling zone
After PCIE controller processing, signal is temporarily stored in inside FIFO, then after AHCI controller and SATA PHY modules,
What is finally exported is exactly SATA signaling interface.
SATA extended chips are that SATA signal is first passed through after PHY modules, some no packets are rejected, simultaneously
Useful data packet analysis is gone out useful data, these data are transferred to DATA FIFO, then the data distribution in FIFO
After 2 SATA PHY modules, unified SATA buses are compressed into.
Embodiment 2：
A kind of network file access method, is to be based on memory cell network big data base frame platform described in embodiment 1
File access method, as shown in figure 1, the platform includes internet 1, be connected to depositing for multiple operators on internet
Storage management server 7 and the terminal user 2 for being registered in different operators, storage managing server pass through the multiple nodes of network connection
Storage server 3, node store server connection data memory cell 4；HASH values are also associated with the internet respectively
Server 5 and user's stored record server 6；Wherein, the network file access method is：When user sets up or updated network
During file, the network file is produced by HASH values server first and a current time HASH value is assigned, while right
The file for assigning current time HASH values sets up file sitation index, the current time in user's stored record server
HASH values are unique newest HASH values, and user is according to newest HASH values extraction document.
Hash also referred to as " is hashed ", is exactly the input random length（It is called and does preliminary mapping, pre-image）, by dissipating
Row algorithm, is transformed into the output of regular length, and the output is exactly hashed value.This conversion is a kind of compression mapping, it is, dissipating
The space of train value is generally much less than the space of input, and different inputs may hash to identical output, it is impossible to from
Hashed value uniquely determines input value.It is exactly briefly a kind of message compression by random length to a certain regular length
The function of eap-message digest.It is file identification server as HASH value servers, when user uploads new file or updates ancient deed,
This document will obtain a new HASH value from HASH value servers, the HASH value old to substitute original, and can broadcast
Which which it is linked to whole system this document storage on the node in operator, user will basis when downloading this document again
What new HASH values were provided, which be linked to, is passed the file of user by backbone network on which node of the operator that opposite face is answered
Come on to the place node of user and download simultaneously, while user downloads file, the storage of current operator
File on node also synchronized update.
Therefore, in embodiment：The user is according to the process of newest HASH values extraction document：When terminal user sends text
After part processing request, the accessing user of user whether which operator, when carrying out storage request, passes through intelligently parsing first
DNS systems are by the storage managing server under current operator where the request analysis of terminal user to user, and user is in note
Volume, log in after set up or open the catalogue file list of oneself, find the file that oneself download or update, click on what is downloaded
File, is found corresponding by this document HASH values with being stored in user's stored record server that internet is connected first
Link, the HASH values of this document are newest if on the node memory where this document link is the operator that user logs in
, then this document will directly download；If on the node for the operator that this document is not logged in user, passing through chain
Connect from backbone network quickly by where the file download of user to user on the node memory of operator, user is then again
Download.
It is the username and password that the registration of user is recorded in the form of database as user's stored record server, also
The HASH values of file are paid in the file link for having user to store with HASH servers.Therefore, in embodiment：It is described to be stored in user
The process of file sitation index is set up in record server is：After user sets up a new file or more new file, lead to first
The HASH values for crossing HASH values server one current time of acquisition being arranged on internet assign new file or update text
Part, by the node memory where the imparting current time HASH values file deposit current operator, to it is described current when
Between HASH values file assign current operator where node chained address；By filename, current time HASH values and chain ground connection
User's stored record server generation sitation index that location deposit is connected with internet, and when filename is identical, update
The HASH values of same file name and chained address.
In embodiment：The node store server is the node store server of ARM frameworks, including ARM frameworks center
Processor；The central processing unit includes PCIE series interfaces circuit and SATA series interfaces circuit and RGM series interfaces
Circuit；The PCI-E series interfaces circuit turns SATA bridge by a PCI-E and connects circuit at least two SATA hard disc data of connection
Memory cell, the SATA series interfaces circuit connects at least two SATA hard disc data storages by a SATA expanded circuit
Unit, the RGM series interfaces circuit is connected through network PHY with internet.
For safe operation, in embodiment：The HASH values server and the family stored record server use two-shipper
Hot standby mode works.
Storage managing server described in embodiment uses intelligent DNS resolution system：The most basic function of DNS policy resolutions
It is the address base on the source IP address used by the user of judgement access website that can be intelligent, contrast dns server and then sensing
On the IP of the node server of operator where user.If visitor is UNICOM user, DNS policy resolutions server can be you
The corresponding UNICOM's IP address of domain name parse and give this visitor.If visitor is telecommunication user DNS policy resolution server meetings
The corresponding telecommunications IP address of your domain name is parsed and gives this visitor.Server is also effectively solved while intelligently parsing
Load balancing.
Therefore, the network file access method further comprises accessing node store server file and monitored；Monitoring
Method be：Pass through SNMP（Simple network agreement）By storage managing server in real time except the hardware performance and work of monitoring device
Make outside state, go back the network traffics of monitoring node server.And alarm and early warning after super warning.That is real-time detection node store
The networking state of server, cpu load, memory size and the utilization rate of monitoring node storage server, memory capacity and utilization
Rate, the network bandwidth and utilization rate, the condition information of hard disk operation, by the information acquired in monitoring, then feeding back to scheduling
System；
When whole system even running, scheduling system is not triggered；
When the network traffics and storage usage amount of some node are more than or equal to the 80% of usage amount, the access of file is adjusted
Spend under other nodes of operator；
When a node does not have handshaking information or delay of response, such as Beijing UNICOM node have continuous 5 minutes not to
The SNMP information or PING that network management system feeds back the machine are obstructed, and system will be automatically by this node from DNS database server system
Deletion of node information in system, makes new request not to be distributed on malfunctioning node；The access of file is dispatched to its of operator
Under his node, until node recovers.Wherein, described operator includes two kinds of situations：
A kind of situation is current operator of the user request end source address based on；For example：Some section of Beijing UNICOM
Point reaches after the early warning value that flow value reaches system that system will be that the request of the user of Beijing UNICOM newly is according to source address automatically
It is transferred on other nodes of Beijing UNICOM；
Another situation is：The operator is that have request end source address by other operators for exporting；Such as Beijing
Some node of UNICOM reached after the early warning value that flow value reaches system, system automatically by be according to source address Beijing UNICOM user
For example middle electricity China of other operators that the outlet of Beijing UNICOM is transferred in new request is logical, and by the middle logical node of electricity China in addition node
When be just defined as scheduling node under great Qu same operator.Similarly there be Shandong UNICOM Henan in Tianjin UNICOM or Hebei UNICOM
Node is led in electricity China during the user of Shanxi UNICOM of UNICOM also calls in when reaching upper limit flow value.It is all actually on network
Node memory.
In embodiment：The method of the monitoring further comprises manual intervention：When a certain node has special demands, for example
During debugging demand, scheduling customer flow can be manually forced to be run to the other node of same operator, at the end of demand by hand
Scheduling is returned.
Claims (6)
1. a kind of network file access method, is the file access side based on memory cell network big data base frame platform
Method, the platform includes being connected to the storage managing server of multiple operators on internet, internet and is registered in not
With the terminal user of operator, storage managing server is connected by network connection node storage server, node store server
Connect data storage cell；HASH values server and user's stored record server are also associated with the internet respectively；Its
It is characterised by, the network file access method is：When user sets up or updates network file, serviced first by HASH values
Device produces to the network file and assigns a current time HASH value, while existing to the file for assigning current time HASH values
File sitation index is set up in user's stored record server, the current time HASH values are unique newest HASH values, user
According to newest HASH values extraction document；
The user is according to the process of newest HASH values extraction document：, will eventually after terminal user sends document processing request
On storage managing server under the request analysis of end subscriber to current operator where user, user builds after registering, logging in
Stand or open the catalogue file list of oneself, find the file that oneself download or update, click on the file to be downloaded, lead to first
Cross with this document HASH values for storing in user's stored record server that internet is connected find it is corresponding link, if this
File link is that then the HASH values of this document are newest on node memory where the operator that user logs in, then this article
Part will directly download；If on the node for the operator that this document is not logged in user, by linking user's
File download is to coming on the node memory of operator where user, and then user downloads again.
2. a kind of network file access method according to claim 1, it is characterised in that described in user's stored record clothes
The process of file sitation index is set up in business device is：After user sets up a new file or more new file, first by setting
The HASH values that HASH values server on internet obtains a current time assign new file or more new file, by institute
State on the node memory where imparting current time HASH values file is stored in current operator, to the current time HASH values
File assigns the chained address of current operator place node；By filename, current time HASH values and chained address deposit with
User's stored record server generation sitation index of internet connection, and when filename is identical, update same file
The HASH values of name and chained address.
3. a kind of network file access method according to claim 1, it is characterised in that the node store server is
The node store server of ARM frameworks, including ARM framework central processing units；The central processing unit includes PCIE series and connect
Mouth circuit and SATA series interfaces circuit and RGM series interfaces circuits；The PCI-E series interfaces circuit passes through a PCI-
E turns SATA bridge and connects circuit at least two SATA hard disc data storage cells of connection, and the SATA series interfaces circuit passes through one
SATA expanded circuits connect at least two SATA hard disc data storage cells, and the RGM series interfaces circuit is through network PHY and mutually
Join network connection.
4. a kind of network file access method according to claim 1 or 2, it is characterised in that the HASH values server
Worked with the family stored record server by the way of two-node cluster hot backup.
5. a kind of network file access method according to claim 1, it is characterised in that the network file access method
Further comprise accessing node store server file and monitor；The method of monitoring is：Storage managing server real-time detection section
The networking state of point storage server, cpu load, memory size and the utilization rate of monitoring node storage server, memory capacity
And utilization rate, the network bandwidth and utilization rate, the condition information of hard disk operation；When the network traffics and storage usage amount of some node
More than or equal to usage amount 80% when, the access of file is dispatched under other nodes of operator；
When a node does not have handshaking information or delay of response, the access of file is dispatched under other nodes of operator, directly
Recover to node.
6. a kind of network file access method according to claim 5, it is characterised in that the operator includes asking
Hold current operator of the source address based on and have request end source address by other operators for exporting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410644353.5A CN104391901B (en) | 2014-11-14 | 2014-11-14 | A kind of memory cell network big data base frame platform and its file access method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410644353.5A CN104391901B (en) | 2014-11-14 | 2014-11-14 | A kind of memory cell network big data base frame platform and its file access method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104391901A CN104391901A (en) | 2015-03-04 |
| CN104391901B true CN104391901B (en) | 2017-07-14 |
Family
ID=52609805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410644353.5A Expired - Fee Related CN104391901B (en) | 2014-11-14 | 2014-11-14 | A kind of memory cell network big data base frame platform and its file access method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104391901B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107085170A (en) * | 2017-06-23 | 2017-08-22 | 盛秀群 | Extra high voltage direct current transmission line based on big data is maked an inspection tour and reponse system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100505762C (en) * | 2006-04-19 | 2009-06-24 | 华中科技大学 | Distributed multi-stage buffer storage system suitable for object network storage |
| CN101252603B (en) * | 2008-04-11 | 2011-03-30 | 清华大学 | Cluster distributed type lock management method based on storage area network SAN |
| CN101645039B (en) * | 2009-06-02 | 2011-06-22 | 中国科学院声学研究所 | Method for storing and reading data based on Peterson graph |
| KR20140011527A (en) * | 2012-06-29 | 2014-01-29 | 한국전자통신연구원 | System for interoperating dht-based overlay networks and method thereof |
| CN103856503B (en) * | 2012-11-29 | 2017-04-12 | 北京华胜天成科技股份有限公司 | Processing method of file data of NAS cluster system and NAS cluster system |
| US9451024B2 (en) * | 2013-03-15 | 2016-09-20 | Futurewei Technologies, Inc. | Self-organizing disk (SoD) |
| CN103942013B (en) * | 2014-04-21 | 2016-09-07 | 北京网视通联科技有限公司 | High-speed read-write and mass-storage system and method for work thereof under a kind of ARM platform |
-
2014
- 2014-11-14 CN CN201410644353.5A patent/CN104391901B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN104391901A (en) | 2015-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10069916B2 (en) | System and method for transparent context aware filtering of data requests | |
| CN110225074B (en) | Communication message distribution system and method based on equipment address domain | |
| CN103152393B (en) | A kind of charging method of cloud computing and charge system | |
| CN103763383B (en) | Integrated cloud storage system and its storage method | |
| CN106709003A (en) | Hadoop-based mass log data processing method | |
| CN104965850B (en) | A kind of database high availability implementation method based on open source technology | |
| CN103116661B (en) | A kind of data processing method of database | |
| CN103870297B (en) | The performance data collection system and method for virtual machine in cloud computing environment | |
| CN103560943B (en) | Network analytic system and method supporting real-time mass data processing | |
| US20110040771A1 (en) | Distributed hardware-based data querying | |
| EP2563062A1 (en) | Long connection management apparatus and link resource management method for long connection communication | |
| CN102413172B (en) | Parallel data sharing method based on cluster technology and apparatus thereof | |
| CN103152352A (en) | Perfect information security and forensics monitoring method and system based on cloud computing environment | |
| CN103645904B (en) | A kind of cache implementing method of interface interchange | |
| CN103312791A (en) | Internet of things heterogeneous data storage method and system | |
| CN109218100A (en) | Distributed objects storage cluster and its request responding method, system and storage medium | |
| EP3285186B1 (en) | Methods and procedures for timestamp-based indexing of items in real-time storage | |
| CN110071978A (en) | A kind of method and device of cluster management | |
| CN103595799A (en) | Method for achieving distributed shared data bank | |
| CN110048896A (en) | A kind of company-data acquisition methods, device and equipment | |
| CN107070752B (en) | Testing method and testing system for long connection capacity | |
| CN103327116A (en) | Dynamic copy storage method for network file | |
| CN104281980B (en) | Thermal power generation unit remote diagnosis method and system based on Distributed Calculation | |
| CN106686099A (en) | Method of realizing active-active mode across machine rooms of OracleRAC database based on infiniband network | |
| CN103561060A (en) | Communication link method under multi-thread environment and transit server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170714 Termination date: 20201114 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |