CN105827744A - Data processing method of cloud storage platform - Google Patents
Data processing method of cloud storage platform Download PDFInfo
- Publication number
- CN105827744A CN105827744A CN201610404870.4A CN201610404870A CN105827744A CN 105827744 A CN105827744 A CN 105827744A CN 201610404870 A CN201610404870 A CN 201610404870A CN 105827744 A CN105827744 A CN 105827744A
- Authority
- CN
- China
- Prior art keywords
- node
- server cluster
- client
- data
- service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/101—Server selection for load balancing based on network conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1029—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers using data related to the state of servers by a load balancer
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a data processing method of a cloud storage platform. The method comprises the steps of before data are written in, determining a storage node through a storage load balance strategy; after the data are written in, re-distributing a copy storage node according to the access frequency or the node storage capacity. By means of the data processing method of the cloud storage platform, the server cluster deployment mode is simplified, a user is prevented from directly operating a server cluster, and reasonability and data stability of the storage nodes are guaranteed.
Description
Technical field
The present invention relates to cloud storage, particularly to the data processing method of a kind of cloud storage platform.
Background technology
Cloud storage have employed the technology such as cloud computing, distributed file system and server cluster, storage resources various in network is aggregating, and common externally offer data storage and Operational Visit function, in scientific research, production and auto service field extensive application.Current cloud storage is divided into publicly-owned service type to store, and i.e. provides storage service to enterprise or individual；One be privately owned architected cloud storage, i.e. enterprises build based on storage server cluster and distributed file system, be deployed in the node trustship place of enterprise data center or safety, provide for enterprise self and store service accordingly.Cloud storage platform confidentiality is higher, and storing process is without too many I/O operation, and therefore using and building private cloud storage system to preserve its data file is best selection.At present, private cloud storage building plan has a variety of: including key assignments type distributed file system, have employed the mode of packet, server cluster is constituted by one or more groups, is the most standby relation with the service node in group.The mode using packet storage can make storage server cluster more flexible, and controllability is the strongest.But, Hadoop is as a distributed storage Computational frame increased income, the shortcoming also having its own.That is exactly system architecture design complexity, and operation maintenance difficulty is bigger.Use to cloud storage platform not only needs many knowledge accumulation, and also has a lot of technical ability to remove learning and mastering in terms of its operation maintenance, and the industry limiting cloud storage platform to a certain extent is promoted and uses.In building information cloud storage platform, also two impacts are disposed and the problem of systematic function: first is in running, and node is susceptible to fault.Once node failure occurs and can not prepare to process in time, will affect in the storage server cluster build process of multiple node, and each node has the operation of a lot of repetition so that build process is the most loaded down with trivial details and easily makes mistakes；Second is because node and is ordinary personal computers, rather than the private server of minicomputer or large scale computer etc, therefore data are in use, and by such as CPU, the impact such as internal memory and magnetic disc i/o is the most serious.
Summary of the invention
For solving the problem existing for above-mentioned prior art, the present invention proposes the data processing method of a kind of cloud storage platform, including:
Determine memory node by storage load balancing strategy before write in data, redistribute copy memory node according to access frequency or node storage capacity after the data writing is finished.
Preferably, the method farther includes: arranges node in the file system of cloud storage platform and selects and dispatch monitor, its interior joint selection strategy is implemented among namenode, called when selecting service node by namenode, dispatch monitor is used for monitoring server cluster operation conditions, and including access frequency and the memory capacity of service node of data block, system is when idle state, management node, according to data access frequency and power system capacity, dispatches copy deposit position；
Before files passe, service node sends write data requests to namenode, namenode calls node preference pattern, by dispatch monitor, obtains server cluster operation information, calculate the stored ratio of node, calculate the stored ratio of each frame node, and according to backup factor number, the node that prioritizing selection stored ratio is the highest forms node queue and is sent to client, by client, data to be stored are divided into multiple data block, are stored on different service nodes；
After All Files has been stored in server cluster, server cluster operation information is collected by dispatch monitor, get data access frequency and the memory capacity of all nodes of all nodes, if the access frequency of data exceedes predefined threshold value, then by Replica placement on the minimum node of access frequency；If system spare capacity is less than threshold value, then by Replica placement at the highest node of stored ratio；
Server cluster operation information is measured by cloud storage platform operation information and display frame monitors, the task that described Framework monitors is relevant with scheduling cloud storage platform；Client selects to be divided into two ways at node: client selection strategy in server cluster node and client selection mode outside server cluster node, and specific implementation is as follows:
Each frame at n rack server cluster arranges TR platform service node, and number of copies is r；If client is on server cluster service node, then
A) client sends write data requests to management node；
B) management node is according to file content and system configuration scenarios, calculates all service node stored ratio of client place frame, and process is as follows:
If client is i-th frame, initialize selected node set SDN for sky；
The residual capacity of this frame jth node is CLij, the block number of storage is BLij, the storage preferred proportion RS of nodeij=CLij/BLij, storage is preferably put into selected node set than two the highest nodes, i.e. SDN={DNia、DNib}；Wherein DNia、DNibRepresent a and the b service node in i-th frame,
C) from remaining the node calculating r-2 storage each frame than maximum, select r-2 node of maximum to put into selected node set SDN after sequence, altogether r node, be used for depositing data block and copy thereof；
D) the node distribution service node during SDN is gathered by management node, to client, is write by client；
When client is not on service node, then the stored ratio of all nodes in direct calculation server cluster, before selecting, r maximum node, is data memory node；FromIn individual node, according to RSij=CLij/BLijSelect r the node that stored ratio is the highest, put in SDN list, be the optimum node chosen.
The present invention compared to existing technology, has the advantage that
The present invention proposes the data processing method of a kind of cloud storage platform, simplifies server cluster deployment way, it is to avoid server cluster is directly operated by user, it is ensured that the reasonability of memory node and data stability.
Accompanying drawing explanation
Fig. 1 is the flow chart of the data processing method of cloud storage platform according to embodiments of the present invention.
Detailed description of the invention
Hereafter provide the detailed description to one or more embodiment of the present invention together with the accompanying drawing of the diagram principle of the invention.Describe the present invention in conjunction with such embodiment, but the invention is not restricted to any embodiment.The scope of the present invention is limited only by the appended claims, and the present invention contains many replacements, amendment and equivalent.Illustrate many details in the following description to provide thorough understanding of the present invention.These details are provided for exemplary purposes, and the present invention can also be realized according to claims without some in these details or all details.
An aspect of of the present present invention provides the data processing method of a kind of cloud storage platform.Fig. 1 is the data processing method flow chart of cloud storage platform according to embodiments of the present invention.
In order to preferably manage server cluster, the present invention carries out automated management to the whole life cycle of cloud storage platform Distributed Architecture running, including installing, builds and monitors, it is provided that visualization interface, improves the efficiency of manager.Storage resource control system carries out fault alarm and process simultaneously.Except the operation maintenance of server cluster is operated, in addition it is also necessary to the performance of server cluster is optimized.In server cluster after newly-increased node, re-optimization performance of server cluster.For cloud storage server cluster variety of problems during deployment, operation maintenance and use, the present invention is directed to the server cluster built, utilize the node scheduling Optimized model in reading and writing data stage, realize convenient management and the optimization of server cluster.The aspect of deployment framework, node administration and server Optimized Operation for server cluster is described in detail by the present invention.
The present invention uses host-guest architecture, comprises a management node and multiple service node.Management node is used for mutual with service node, the heartbeat request that node of accepting business sends, and completes centralized management watchdog logic, and each service node is responsible for state acquisition and the maintenance work of place node.Management node deployment is at single node, management node as server cluster deployment framework, its responsibility is that the order receiving user's transmission performs request, order is sent to service node with rear, using JSON mode to send order, these JSON data include installation, start, stop the configuration information of service.
Service node is deployed on the node of all server clusters to be added, it is used for the execution task requests performing to be sent by management node, performed script is stored under the assigned catalogue on management node, the content transformation of the command file that service node receives Self management node is dictionary format by this script, it is simple to the use configured when script realizes disposing.Dispose during state and behavior transmission all for by management node be sent to service node, service node receives certain operation behavior, by behavior perform thread perform correspondence method, and will execution after message feed back to management node by message queue.
During server cluster is disposed, operator perform different behaviors by the page, and management node will be sent to service node the behavior. performed thread by the behavior of service node again and perform the operation of correspondence, complete server cluster and dispose.During service node performs, send back to the status information in server cluster manage node, the finite state machine of management node judge.
In server cluster node configures, the present invention gives tacit consent to all nodes and the most successfully installs operating system, either physical machine or virtual machine.Node joins two steps in server cluster, first is both sides' safety certifications, and second is the configuration of node name.Both sides' safety certification uses Shell script to write, and system performs this script, by the PKI file distributing of management node to each service node, to reach the state without password login.
Configuration service device cluster service includes selecting service and selecting service place node.Current information on services writes in JSON data, when selecting service by reading this JSON data, obtains all of cloud storage platform service, optionally installs.After selecting service, distribute at corresponding node, the node listing before at this moment reading by service, then each service is carried out node selection.
Service and node configuration information all oneself after setting completed, corresponding cloud storage platform service installation kit can be distributed on the node of correspondence by performing Shell, and install by system.The service profile information of all nodes synchronizes.
After server cluster has been built, server cluster interior joint increases along with the increase of data volume, and the complexity of the interpolation of node and the process of malfunctioning node the most exponentially goes up.The present invention utilizes node administration, monitors thread by a cloud storage platform service, is polled the cloud storage platform service of management node, monitors the running status of each node on server cluster in real time, realizes increasing node and deletion action simultaneously.
Node administration uses observer's pattern to realize cloud storage platform service monitor, and wherein node manager is observer, and cloud storage platform service monitor is the person of being observed.Supervision includes monitoring server cluster operation conditions, including all node health, file system service condition.Management includes server cluster node and the unlatching of service, closedown, the increase of node and deletion etc..
After cloud storage Platform Server is built, management node can start server cluster and monitor, periodically initiates nodal information and obtains server cluster node index.Selected node to increase or the operation of deletion by cloud storage platform nodes manager simultaneously.
If there being new node server cluster to be added, cloud storage platform nodes manager is then needed to obtain the nodal information of server cluster to be added, send that information to cloud storage platform nodes monitor again, by performing relevant Shell order, judge whether this node can join server cluster, and feedback-related information is to cloud storage platform nodes manager, cloud storage platform nodes manager select the concrete behavior of node.In like manner, knot removal is also required to first from cloud storage platform nodes manager, and the nodal information that will delete is sent to cloud storage platform nodes monitor, and monitor judges the state of this node, performs node and removes operation.
The server optimization that the present invention proposes includes that node selects and storage scheduling.Node selects to refer to that data strategically specify memory node by some before write, ensure that storage load balance, after storage scheduling refers to data write, redistribute copy memory node according to access frequency or node storage capacity so that system can be run more efficiently.
File system is provided with node and selects and dispatch monitor.Its interior joint selection strategy is implemented among namenode, namenode call when selecting service node.Dispatch monitor is used for monitoring server cluster operation conditions, including access frequency and the memory capacity of service node of data block, system is when idle state, and management node is according to data access frequency and power system capacity, scheduling copy deposit position, improves running efficiency of system.
Node selects to specifically include, before files passe, service node sends write data requests to namenode, namenode calls node preference pattern, pass through dispatch monitor, obtain server cluster operation information, calculate the stored ratio of node, calculate the stored ratio of each frame node, and according to backup factor number, the node that prioritizing selection stored ratio is the highest forms node queue, is sent to client, by client, data to be stored are divided into multiple data block, are stored on different service nodes.
Storage scheduling specifically includes, after All Files has been stored in server cluster, server cluster operation information is collected by dispatch monitor, get data access frequency and the memory capacity of all nodes of all nodes, when server cluster is in idle state, management node, according to the data got, carries out storage scheduling.If the access frequency of data exceedes predefined threshold value, then by Replica placement on the minimum node of access frequency；If system spare capacity is less than threshold value, then by Replica placement at the highest node of stored ratio.
Below for server optimization scheduling model, its monitor and optimisation strategy are described.
First, server cluster operation information needs to be measured by cloud storage platform operation information to monitor with display frame, survey tool is extended by described framework, it is provided that can show in real time and the instrument of historical data for one, help the task that monitoring is relevant with scheduling cloud storage platform.
Cloud storage scheduling in the present invention is divided into two stages.Wherein first stage is that service node during file write selects, and is to be realized by optimization node selection strategy.Client selects to be divided into two ways at node: client selection strategy in server cluster node and client selection mode outside server cluster node.Implementation is as follows:
Assuming that server cluster has n frame, each frame to have TR platform service node, number of copies is r
If client is on server cluster service node, then
A) client sends write data requests to management node；
B) management node is according to file content and system configuration scenarios, calculates all service node stored ratio of client place frame, and process is as follows
If client is i-th frame, initialize selected node set SDN for sky；
The residual capacity of this frame jth node is CLij, the block number of storage is BLij, the storage preferred proportion RS of nodeij=CLij/BLij, storage is preferably put into selected node set than two the highest nodes, i.e. SDN={DNia、DNib}；Wherein DNia、DNibRepresent a and the b service node in i-th frame,
C) from remaining the node calculating r-2 storage each frame than maximum, select r-2 node of maximum to put into selected node set SDN after sequence, altogether r node, be used for depositing data block and copy thereof；
D) the node distribution service node during SDN is gathered by management node, to client, is write by client.
When client is not on service node, then the stored ratio of all nodes in direct calculation server cluster, before selecting, r maximum node, is data memory node.FromIn individual node, according to RSij=CLij/BLijSelect r the node that stored ratio is the highest, put in SDN list, be the optimum node chosen.
In sum, the present invention proposes the data processing method of a kind of cloud storage platform, simplifies server cluster deployment way, it is to avoid server cluster is directly operated by user, it is ensured that the reasonability of memory node and data stability.
Obviously, it should be appreciated by those skilled in the art, each module of the above-mentioned present invention or each step can realize by general calculating system, they can concentrate in single calculating system, or being distributed on the network that multiple calculating system is formed, alternatively, they can realize with the executable program code of calculating system, it is thus possible to be stored in storage system being performed by calculating system.So, the present invention is not restricted to the combination of any specific hardware and software.
It should be appreciated that the above-mentioned detailed description of the invention of the present invention is used only for exemplary illustration or explains the principle of the present invention, and it is not construed as limiting the invention.Therefore, any modification, equivalent substitution and improvement etc. done in the case of without departing from the spirit and scope of the present invention, should be included within the scope of the present invention.Additionally, claims of the present invention be intended to fall in the equivalents on scope and border or this scope and border whole change and modifications example.
Claims (2)
1. the data processing method of a cloud storage platform, it is characterised in that including:
Determine memory node by storage load balancing strategy before write in data, redistribute copy memory node according to access frequency or node storage capacity after the data writing is finished.
Method the most according to claim 1, it is characterized in that, the method farther includes: arranges node in the file system of cloud storage platform and selects and dispatch monitor, its interior joint selection strategy is implemented among namenode, called when selecting service node by namenode, dispatch monitor is used for monitoring server cluster operation conditions, access frequency and the memory capacity of service node including data block, system is when idle state, management node, according to data access frequency and power system capacity, dispatches copy deposit position；
Before files passe, service node sends write data requests to namenode, namenode calls node preference pattern, by dispatch monitor, obtains server cluster operation information, calculate the stored ratio of node, calculate the stored ratio of each frame node, and according to backup factor number, the node that prioritizing selection stored ratio is the highest forms node queue and is sent to client, by client, data to be stored are divided into multiple data block, are stored on different service nodes；
After All Files has been stored in server cluster, server cluster operation information is collected by dispatch monitor, get data access frequency and the memory capacity of all nodes of all nodes, if the access frequency of data exceedes predefined threshold value, then by Replica placement on the minimum node of access frequency；If system spare capacity is less than threshold value, then by Replica placement at the highest node of stored ratio；
Server cluster operation information is measured by cloud storage platform operation information and display frame monitors, the task that described Framework monitors is relevant with scheduling cloud storage platform；Client selects to be divided into two ways at node: client selection strategy in server cluster node and client selection mode outside server cluster node, and specific implementation is as follows:
Each frame at n rack server cluster arranges TR platform service node, and number of copies is r；If client is on server cluster service node, then
A) client sends write data requests to management node；
B) management node is according to file content and system configuration scenarios, calculates all service node stored ratio of client place frame, and process is as follows:
If client is i-th frame, initialize selected node set SDN for sky；
The residual capacity of this frame jth node is CLij, the block number of storage is BLij, the storage preferred proportion RS of nodeij=CLij/BLij, storage is preferably put into selected node set than two the highest nodes, i.e. SDN={DNia、DNib}；Wherein DNia、DNibRepresent a and the b service node in i-th frame,
C) from remaining the node calculating r-2 storage each frame than maximum, select r-2 node of maximum to put into selected node set SDN after sequence, altogether r node, be used for depositing data block and copy thereof；
D) the node distribution service node during SDN is gathered by management node, to client, is write by client；
When client is not on service node, then the stored ratio of all nodes in direct calculation server cluster, before selecting, r maximum node, is data memory node；FromIn individual node, according to RSij=CLij/BLijSelect r the node that stored ratio is the highest, put in SDN list, be the optimum node chosen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610404870.4A CN105827744A (en) | 2016-06-08 | 2016-06-08 | Data processing method of cloud storage platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610404870.4A CN105827744A (en) | 2016-06-08 | 2016-06-08 | Data processing method of cloud storage platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105827744A true CN105827744A (en) | 2016-08-03 |
Family
ID=56532164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610404870.4A Pending CN105827744A (en) | 2016-06-08 | 2016-06-08 | Data processing method of cloud storage platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105827744A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106302656A (en) * | 2016-08-01 | 2017-01-04 | 成都鼎智汇科技有限公司 | The Medical Data processing method of cloud storage platform |
| CN106453650A (en) * | 2016-11-30 | 2017-02-22 | 安徽金曦网络科技股份有限公司 | Cloud storage system |
| CN110636091A (en) * | 2018-06-22 | 2019-12-31 | 北京东土科技股份有限公司 | Data balancing method, device, equipment and storage medium for cloud storage cluster |
| CN115174580A (en) * | 2022-09-05 | 2022-10-11 | 睿至科技集团有限公司 | Data processing method and system based on big data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| CN101370030A (en) * | 2008-09-24 | 2009-02-18 | 东南大学 | Resource load stabilization method based on contents duplication |
| CN101686262A (en) * | 2009-05-14 | 2010-03-31 | 南京大学 | Multi-node collaboration based storage method for sensor network |
| CN102035884A (en) * | 2010-12-03 | 2011-04-27 | 华中科技大学 | Cloud storage system and data deployment method thereof |
| CN103139302A (en) * | 2013-02-07 | 2013-06-05 | 浙江大学 | Real-time copy scheduling method considering load balancing |
| CN103150347A (en) * | 2013-02-07 | 2013-06-12 | 浙江大学 | Dynamic replica management method based on file heat |
-
2016
- 2016-06-08 CN CN201610404870.4A patent/CN105827744A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| CN101370030A (en) * | 2008-09-24 | 2009-02-18 | 东南大学 | Resource load stabilization method based on contents duplication |
| CN101686262A (en) * | 2009-05-14 | 2010-03-31 | 南京大学 | Multi-node collaboration based storage method for sensor network |
| CN102035884A (en) * | 2010-12-03 | 2011-04-27 | 华中科技大学 | Cloud storage system and data deployment method thereof |
| CN103139302A (en) * | 2013-02-07 | 2013-06-05 | 浙江大学 | Real-time copy scheduling method considering load balancing |
| CN103150347A (en) * | 2013-02-07 | 2013-06-12 | 浙江大学 | Dynamic replica management method based on file heat |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106302656A (en) * | 2016-08-01 | 2017-01-04 | 成都鼎智汇科技有限公司 | The Medical Data processing method of cloud storage platform |
| CN106453650A (en) * | 2016-11-30 | 2017-02-22 | 安徽金曦网络科技股份有限公司 | Cloud storage system |
| CN110636091A (en) * | 2018-06-22 | 2019-12-31 | 北京东土科技股份有限公司 | Data balancing method, device, equipment and storage medium for cloud storage cluster |
| CN115174580A (en) * | 2022-09-05 | 2022-10-11 | 睿至科技集团有限公司 | Data processing method and system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106101213A (en) | Information-distribution type storage method | |
| Hu et al. | Flutter: Scheduling tasks closer to data across geo-distributed datacenters | |
| CN112685170B (en) | Dynamic optimization of backup strategies | |
| CN107567696A (en) | The automatic extension of resource instances group in computing cluster | |
| EP4121856B1 (en) | Systems, methods, computing platforms, and storage media for administering a distributed edge computing system utilizing an adaptive edge engine | |
| Araujo et al. | Availability evaluation of digital library cloud services | |
| CN106101212A (en) | Big data access method under cloud platform | |
| US8381222B2 (en) | Policy driven automation—specifying equivalent resources | |
| Teng et al. | Simmapreduce: A simulator for modeling mapreduce framework | |
| CN105359147A (en) | Online database migration | |
| Zhang et al. | Improving Hadoop service provisioning in a geographically distributed cloud | |
| CN109614227A (en) | Task resource concocting method, device, electronic equipment and computer-readable medium | |
| Debski et al. | A scalable, reactive architecture for cloud applications | |
| Lebre et al. | Putting the next 500 vm placement algorithms to the acid test: The infrastructure provider viewpoint | |
| Patni et al. | Load balancing strategies for grid computing | |
| Bermbach et al. | On the future of cloud engineering | |
| CN105827744A (en) | Data processing method of cloud storage platform | |
| US20120246318A1 (en) | Resource compatability for data centers | |
| Deng et al. | A clustering based coscheduling strategy for efficient scientific workflow execution in cloud computing | |
| Vaquero et al. | Deploying large-scale datasets on-demand in the cloud: treats and tricks on data distribution | |
| CN106254452A (en) | The big data access method of medical treatment under cloud platform | |
| Xu et al. | Fault tolerance and quality of service aware virtual machine scheduling algorithm in cloud data centers | |
| CN106302656A (en) | The Medical Data processing method of cloud storage platform | |
| Jain et al. | Cloud service orchestration based architecture of OpenStack Nova and Swift | |
| Bellavista et al. | GAMESH: a grid architecture for scalable monitoring and enhanced dependable job scheduling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160803 |