CN102110041A - System and method for monitoring storage space of distributed file system - Google Patents
System and method for monitoring storage space of distributed file system Download PDFInfo
- Publication number
- CN102110041A CN102110041A CN2011100610371A CN201110061037A CN102110041A CN 102110041 A CN102110041 A CN 102110041A CN 2011100610371 A CN2011100610371 A CN 2011100610371A CN 201110061037 A CN201110061037 A CN 201110061037A CN 102110041 A CN102110041 A CN 102110041A
- Authority
- CN
- China
- Prior art keywords
- namespace
- server
- book
- file system
- distributed file
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000013507 mapping Methods 0.000 claims description 20
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000003247 decreasing effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000013209 evaluation strategy Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a system and method for monitoring the storage space of a distributed file system. The system comprises at least one or more client servers for continuously monitoring the state of a name space of the distributed file system mapped to the server, and recording the monitored state information, and one or more data servers for providing the name space of the distributed file system to the client server. According to the invention, the client servers can timely upload the monitored state information of the name space of the distributed file system to allow a control-side server to monitor the scheduling and the use of all name spaces of the distributed file system, thereby reliably executing the application business deployed in the name space.
Description
Technical field
The present invention relates to the monitoring of mass storage file system storage space, relate in particular to the system and method that the distributed file system storage space is monitored.
Background technology
Rise along with cloud computing, data processing, files in stream media storage such as high-performance calculation, grid computing, high-energy physics experiment are more and more higher to the capacity and the performance requirement of storage system with the contour performance applications of visit, physical characteristics restriction under the pressure of storage medium, the development of mass-storage system is developed to the distributed storage mode gradually by the extended mode that expands disk, array, distributed file system as the distributed mass memory system core has obtained development faster, becomes mainstream applications gradually.
The control end server provides storage space with the form of distributed file system NameSpace to client-server, and client-server is carried out actual applied business according to mapped distributed file system NameSpace.And the NameSpace of a distributed file system may shine upon simultaneously to different client-servers.
In client-server, be based upon the correct execution of the applied business in the distributed file system NameSpace, depend on the upstate of NameSpace morely, client-server but can not be known the availability (comprising local availability and overall usability) of distributed file system NameSpace in advance on the other hand, and because of the control end server to the adjustment of the volume resource in the NameSpace to the influence that the book availability produces, cause in the client-server cluster environment, disposing a large amount of, tend to lose a large amount of physical resources and financial resources in the scene of important application business.
Distributed file system is in the process that mass memory capacity, extensibility framework and high availability are provided, various problems such as dish, low-quality disk, data are inconsistent, book interpolation, deletion, migration also can appear falling in the book of global name space inevitably, the appearance of these problems all can have influence on the local availability of distributed file system global name space, sometimes even can have influence on its overall usability, when the scale of distributed file system enlarged, it is more serious that the appearance of this class problem can become.
Present evaluation method at distributed file system NameSpace availability, the main monitoring that relies on the control end server of distributed file system to book on all data servers, and introduce client-server book state on the data server of the distributed file system of its carry is not monitored, and there is certain property delayed in time in the usability status monitoring of book as the control end server of main controlled node; In addition, the inefficacy (being local availability) of individual data volume not necessarily can influence the overall usability of distributed file system in the distributed file system, that is the control end server is to the monitoring of book on all data servers none comprehensive evaluation strategy still.
Therefore, need be improved at the evaluation method of existing distributed file system name space availability, make client-server can real-time monitor the status information of mapped NameSpace, and in time offer the control end server, it can in time be monitored all distributed file system NameSpaces, thereby guarantee that all applied business that are deployed in NameSpace carry out reliably.
Summary of the invention
Technical matters to be solved by this invention provides a kind of method and system that realize the monitoring of distributed file system storage space, makes client-server can in time monitor the state of the storage space of mapping.
In order to solve the problems of the technologies described above, the invention provides a kind of realization distributed file system storage space supervisory system, comprise at least: one or more data servers of one or more client-servers, distributed file system, wherein:
Client-server is used for continuing carrying out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors;
Data server is used for providing the distributed file system NameSpace to client-server.
Further,
Data server provides NameSpace with the form of book to client-server;
Client-server is given the quantity of the distributed file system NameSpace of book server by the regular query mappings of life space surveillance passage, check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the described book of record check.
Further, this supervisory system also comprises the control end server, wherein:
Client-server also is uploaded to the control end server with the availability status information of book of record;
The control end server is used for the availability status information of the distributed file system control volume availability status information according to monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
Further, the control of control end server monitoring volume availability status information comprise that the control volume is available, disabled status information any one;
When the control end server was unavailable at the state that monitors control volume, it was unavailable to adjudicate corresponding distributed system NameSpace; But be the time spent at the state that monitors the control volume, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding distributed system NameSpace, then take in the mode decision scheme of following setting one or more to adjudicate the usability status of NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
Further, also be included in selected in a client-server primary client server,
This primary client server is except the book availability status information of monitoring mapping to the distributed file system NameSpace of book server, also collect the book availability status information of the distributed file system NameSpace of other client-server monitoring, and the book availability status information that data monitored is rolled up availability status information and collection is uploaded to the control end server.
Further,
Described NameSpace is dispatched and used to the control end server also according to the availability status information to all distributed file system NameSpaces, the execution of applied business in the monitoring NameSpace.
In order to solve the problems of the technologies described above, the invention provides a kind of client-server in foregoing supervisory system, comprise built-in monitoring modular, this monitoring modular is used for continuing carrying out status monitoring for the distributed file system NameSpace of this client-server to mapping, and writes down the status information that monitors.
Further
The quantity that this monitoring modular is regularly inquired about the distributed file system NameSpace by life space surveillance passage, and check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the book of record is uploaded.
Further, selected primary client server in client-server, wherein:
This primary client server by built-in monitoring modular except the book availability status information of distributed file system NameSpace of book server is given in monitoring mapping, also collect the book availability status information of the distributed file system NameSpace of other client-server monitoring, and the book availability status information that data monitored is rolled up availability status information and collection is uploaded to this control end server.
In order to solve the problems of the technologies described above, the invention provides a kind of control end server in foregoing supervisory system, comprising built-in monitoring module, this monitoring module is used for the availability status information of the distributed file system control volume availability status information according to monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
Further, distributed file system control volume availability status information comprise that the control volume is available, disabled status information any one;
When this monitoring module was unavailable at the state that monitors control volume, it was unavailable to adjudicate corresponding distributed system NameSpace; But be the time spent at the state that monitors the control volume, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding NameSpace, then take in the following predefined mode decision scheme one or more to adjudicate the usability status of described NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
Further,
Monitoring module is also according to availability status information scheduling and use NameSpace to all distributed file system NameSpaces, the execution of applied business in the monitoring NameSpace.
In order to solve the problems of the technologies described above, the invention provides a kind of realization distributed file system storage space method for supervising, relate to one or more client-servers at least, this method comprises:
Client-server continues to carry out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors.
Further, client-server continues to carry out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors, and specifically comprises:
Give the quantity of the distributed file system NameSpace of book server by the regular query mappings of life space surveillance passage, and check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the described book of record check.
Further, this method for supervising also relates to the control end server, and this method also comprises:
Client-server is uploaded to the control end server with the availability status information of the book of record;
The control end server is according to the availability status information of the distributed file system control volume availability status information of monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
Further, the control end server specifically comprises according to the availability status information of the control volume availability status information of monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces:
When the state that monitors control volume when being unavailable, it is unavailable to adjudicate corresponding distributed system NameSpace;
When but the state that monitors the control volume is the time spent, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding NameSpace, then described control end server adopts in the mode decision scheme of following setting one or more to adjudicate the usability status of described NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
Further, this method for supervising also comprises:
The control end server is according to availability status information scheduling and use NameSpace to all distributed file system NameSpaces, the execution of applied business in the monitoring NameSpace.
The present invention is directed to the existing distributed file system method of the NameSpace state being carried out dynamic monitoring and monitoring is provided, make client-server the status information of the distributed file system NameSpace that monitors can be uploaded to the control end server in time, the control end server is according to the monitoring of control volume availability, information that client-server is uploaded and predetermined judge policy information, monitor the scheduling and the use of all distributed file system NameSpaces, carry out reliably thereby assurance is deployed in the applied business of NameSpace.
Description of drawings
Fig. 1 is the system embodiment structural representation of realization distributed file system storage space monitoring of the present invention;
Fig. 2 is that client-server of the present invention is to distributed file system storage space state monitoring method embodiment process flow diagram;
Fig. 3 is that control end server of the present invention is to distributed file system storage space state monitoring method embodiment process flow diagram.
Embodiment
Below in conjunction with accompanying drawing and preferred embodiment technical scheme of the present invention is at length set forth.The embodiment that below exemplifies only is used for description and interpretation the present invention, and does not constitute the restriction to technical solution of the present invention.
The system embodiment of realization distributed file system storage space monitoring provided by the invention, its structure comprises as shown in Figure 1 at least: one or more data servers of one or more client-servers, distributed file system, wherein:
Client-server is used for continuing carrying out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors;
Data server is used for providing the distributed file system NameSpace to client-server.
Data server provides NameSpace with the form of book to client-server;
The built-in monitoring modular of client-server passes through life space surveillance passage, regularly give the quantity of the distributed file system NameSpace of book server according to the mapping of inquiry, check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of record data volume.
System embodiment shown in Figure 1 also comprises the control end server, wherein:
Client-server also is uploaded to the control end server with the availability status information of book of record;
The control end server is used for the availability status information of the distributed file system control volume availability information according to monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
Wherein, the distributed file system of monitoring control volume availability information comprises available, the disabled status information of control volume.
Wherein, the built-in monitoring module of control end server is monitoring the control volume when unavailable, and it is unavailable to adjudicate pairing distributed system NameSpace;
But monitoring the control volume time spent, supposing that the useful number of book in its corresponding NameSpace is that x, unavailable number are that y and abnormal number are z; Then the predefined mode decision scheme of control end server comprise in the following strategy any one or multiple:
(1) if book can be used ratio m in the NameSpace
1=x/ (x+y+z) adjudicates this NameSpace for whole or local available more than or equal to the available ratio threshold of setting (can a more than threshold);
(2) as if the unusual ratio m of book in the NameSpace
2=z/ (x+y+z) is more than or equal to the unusual ratio threshold of setting (can a more than threshold), and adjudicating this NameSpace is integral body or local anomaly.
(3) as if the unavailable ratio m of book in the NameSpace
3=y/ (x+y+z) adjudicates this NameSpace for whole or local unavailable more than or equal to the unavailable ratio threshold of setting (can a more than threshold).
When the client-server number more for a long time, under the situation about need divide into groups to the monitoring of a plurality of client-servers, can in a plurality of client-servers, select a primary client server, this primary client server by built-in monitoring modular except the book usability status of distributed file system NameSpace of book server is given in monitoring mapping, also collect the book availability status information of the distributed file system NameSpace of other client-server monitoring, and the book availability status information that data monitored is rolled up availability status information and collection is uploaded to the control end server.
Because the running status of the single client-server under cloud computing environment is very unreliable, can influence the reliability of its corresponding distributed file system name space book status monitoring; And owing to a distributed file system can be shone upon simultaneously to different client-servers, therefore the present invention is merged the status monitoring of a plurality of client-servers of distributed file system to book in the same NameSpace by control end server or primary client server, can increase objectivity, reliability and the accuracy of status monitoring greatly.
The built-in monitoring module of control end server according to the judgement scheduling of the usability status of all distributed file system NameSpaces with use NameSpace, is to carry out, suspend to carry out and recover and carry out in order to the applied business of determining corresponding NameSpace also.
The present invention is directed to said system embodiment, correspondingly also provide client-server to distributed file system storage space state monitoring method embodiment, its flow process is that client-server is then carried out at predetermined monitoring time as shown in Figure 2, comprises the steps:
110: inquiry distributed file system NameSpace quantity;
120: judging whether to exist NameSpace, is then to carry out the following step, otherwise process ends;
NameSpace quantity according to inquiry judges whether to exist NameSpace.
130: the quantity of inquiry NameSpace book;
140: judging whether to exist book, is then to carry out the following step, otherwise execution in step 170;
150: monitor the state of this book, and record data volume state information;
The book status information of record comprises that book can be with in, the unavailable and abnormality information one or more.
The main control server end has all given the judgement of corresponding local availability timely.
160: the quantity of book is successively decreased, and returns step 140 and carries out;
170: NameSpace quantity is successively decreased;
180: judging whether to finish all NameSpace status monitorings, is execution in step 190 then, carries out otherwise return step 130;
Non-vanishing according to NameSpace quantity, judge and not finish as yet and all NameSpace status monitorings.
190: return data volume state information, process ends.
The book status information of all NameSpaces of return recording.
The present invention is directed to said system embodiment, correspondingly also provide the control end server to distributed file system storage space method for supervising embodiment, its flow process comprises the steps: as shown in Figure 3
210,220: check the control volume state, and whether the control volume is available, is then to carry out the following step, otherwise execution in step 280;
230: the NameSpace quantity of inquiring about all distributed file systems;
240: judging whether to exist NameSpace, is then to carry out the following step, otherwise execution in step 290;
250: the book status information of inquiring about this NameSpace that each client-server uploads;
The present invention is merged the mode that a plurality of client-servers carry out the availability monitoring to identical file systematic naming method space by the control end server, can estimate the overall usability and the local usability status of book in this NameSpace more objectively.
260: according to the book status information and the predefined mode decision scheme availability of determining this NameSpace of inquiry;
Address before the predefined mode decision scheme, this repeats no more.
270: NameSpace quantity is successively decreased, and returns step 240 and carries out;
280: return the availability information of all distributed file system NameSpaces, process ends.
Said method embodiment also comprises:
The control end server is according to dispatching and use NameSpace to the availability information of all distributed file system NameSpaces, and the applied business of determining corresponding NameSpace is that execution, time-out are carried out and recovery is carried out.
The present invention is in the availability monitor of control end server implementation NameSpace curls inward, can guarantee the policer operation of NameSpace availability in the bookkeeping of NameSpace and the client-server is coordinated to carry out.
The present invention carries out following tests to validity, promptness and the accuracy of corresponding system and method after said system and method embodiment are provided:
The distributed file system NameSpace of single client-server is destroyed the disturbed test of the configuration of book and accessibility (readable and/or can write), test corresponding client-server respectively, whether the control end server can in time be judged at the whole or local availability of book;
The control end server initiatively manages operation (for example add, delete or the change operation) to the distributed file system NameSpace, and whether the distributed file system of test control end server can provide unnecessary NameSpace change in availability information;
Destroy the overall usability of distributed file system NameSpace, whether the test control end server can provide corresponding change in availability information at this NameSpace.
Test result shows, said system provided by the invention and method embodiment can make client-server in time the availability information of the book that monitors is uploaded to the control end server, and the control end server is according to scheduling and use that the monitoring of control volume availability, book availability information that client-server is uploaded and predetermined judge policy information are monitored all distributed file system NameSpaces effectively.
More than explanation only is the preferable embodiment of the present invention; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses, the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (17)
1. realize distributed file system storage space supervisory system for one kind, comprise at least: one or more data servers of one or more client-servers, distributed file system is characterized in that:
Client-server is used for continuing carrying out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors;
Data server is used for providing the distributed file system NameSpace to client-server.
2. according to the described system of claim 1, it is characterized in that,
Described data server provides NameSpace with the form of book to described client-server;
Described client-server is given the quantity of the distributed file system NameSpace of book server by the regular query mappings of life space surveillance passage, check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the described book of record check.
3. according to the described supervisory system of claim 2, it is characterized in that, also comprise the control end server, wherein:
Described client-server also is uploaded to the control end server with the availability status information of book of record;
The control end server is used for the availability status information of the book uploaded according to the distributed file system control volume availability status information of monitoring, described client-server and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
4. according to the described supervisory system of claim 3, it is characterized in that, the described control volume availability status information of described control end server monitoring comprises that the control volume is available, disabled status information any one;
When described control end server was unavailable at the state that monitors described control volume, it was unavailable to adjudicate corresponding distributed system NameSpace; But be the time spent at the state that monitors described control volume, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding distributed system NameSpace, then take in the mode decision scheme of following setting one or more to adjudicate the usability status of described NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
5. according to the described supervisory system of claim 3, it is characterized in that, also be included in selected in a described client-server primary client server,
This primary client server is except the book availability status information of monitoring mapping to the distributed file system NameSpace of book server, also collect the book availability status information of the distributed file system NameSpace of other client-server monitoring, and the book availability status information that data monitored is rolled up availability status information and collection is uploaded to described control end server.
6. according to each described supervisory system of claim 3 to 5, it is characterized in that,
Described control end server also according to the availability status information of all distributed file system NameSpaces being dispatched and using described NameSpace, is monitored the execution of applied business in the described NameSpace.
7. client-server in supervisory system as claimed in claim 1, it is characterized in that comprising built-in monitoring modular, described monitoring modular is used for continuing carrying out status monitoring for the distributed file system NameSpace of this client-server to mapping, and writes down the status information that monitors.
8. according to the described client-server of claim 7, it is characterized in that,
The quantity that described monitoring modular is regularly inquired about described distributed file system NameSpace by life space surveillance passage, and check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the described book that will write down is uploaded.
9. according to the described client-server of claim 8, it is characterized in that, selected primary client server in described client-server, wherein:
This primary client server by built-in monitoring modular except the book availability status information of distributed file system NameSpace of book server is given in monitoring mapping, also collect the book availability status information of the distributed file system NameSpace of other client-server monitoring, and the book availability status information that data monitored is rolled up availability status information and collection is uploaded to this control end server.
10. control end server in supervisory system as claimed in claim 3, it is characterized in that comprising built-in monitoring module, described monitoring module is used for the availability status information of the distributed file system control volume availability status information according to monitoring, book that client-server is uploaded and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
11. according to the described control end server of claim 10, it is characterized in that, described distributed file system control volume availability status information comprises that the control volume is available, disabled status information any one;
When described monitoring module was unavailable at the state that monitors described control volume, it was unavailable to adjudicate corresponding distributed system NameSpace; But be the time spent at the state that monitors described control volume, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding NameSpace, then take in the mode decision scheme of following setting one or more to adjudicate the usability status of described NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
12. according to claim 10 or 11 described control end servers, it is characterized in that,
Described monitoring module also according to the availability status information of all distributed file system NameSpaces being dispatched and using described NameSpace, is monitored the execution of applied business in the described NameSpace.
13. realize distributed file system storage space method for supervising for one kind, relate to one or more client-servers at least, it is characterized in that this method comprises:
Described client-server continues to carry out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors.
14., it is characterized in that described client-server continues to carry out status monitoring for the distributed file system NameSpace of book server to mapping, and writes down the status information that monitors, and specifically comprises according to the described method for supervising of claim 13:
Give the quantity of the distributed file system NameSpace of book server by the regular query mappings of life space surveillance passage, and check the usability status of all book in each NameSpace, comprise that book can be with, one or more states in unavailable and unusual, and the availability status information of the described book of record check.
15., it is characterized in that also relate to the control end server, this method also comprises according to the described method for supervising of claim 14:
Described client-server is uploaded to described control end server with the availability status information of the book of record;
The availability status information of the book that described control end server is uploaded according to the distributed file system control volume availability status information of monitoring, described client-server and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces.
16. according to the described method for supervising of claim 15, it is characterized in that, the availability status information of the book that described control end server is uploaded according to the control volume availability status information of monitoring, described client-server and the usability status that predefined mode decision scheme is adjudicated all distributed file system NameSpaces specifically comprise:
When the state that monitors described control volume when being unavailable, it is unavailable to adjudicate corresponding distributed system NameSpace;
When but the state that monitors described control volume is the time spent, if the useful number of book is that x, unavailable number are that y and abnormal number are z in the corresponding NameSpace, then described control end server adopts in the mode decision scheme of following setting one or more to adjudicate the usability status of described NameSpace:
If book can be x/ (x+y+z) with ratio in this NameSpace,, adjudicate this NameSpace for whole or local available when the available ratio threshold of this available ratio more than or equal to setting;
If the unusual ratio of book is z/ (x+y+z) in this NameSpace, when the unusual ratio threshold of this unusual ratio more than or equal to setting, adjudicating this NameSpace is integral body or local anomaly;
If the unavailable ratio of book is y/ (x+y+z) in this NameSpace,, adjudicate this NameSpace for whole or local unavailable when the unavailable ratio threshold of this unavailable ratio more than or equal to setting.
17. according to claim 15 or 16 described method for supervising, it is characterized in that, also comprise:
Described control end server is monitored the execution of applied business in the described NameSpace according to the availability status information of all distributed file system NameSpaces being dispatched and using described NameSpace.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110061037 CN102110041B (en) | 2011-03-15 | 2011-03-15 | System and method for monitoring storage space of distributed file system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110061037 CN102110041B (en) | 2011-03-15 | 2011-03-15 | System and method for monitoring storage space of distributed file system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102110041A true CN102110041A (en) | 2011-06-29 |
CN102110041B CN102110041B (en) | 2013-07-10 |
Family
ID=44174208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201110061037 Active CN102110041B (en) | 2011-03-15 | 2011-03-15 | System and method for monitoring storage space of distributed file system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102110041B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102638378A (en) * | 2012-02-22 | 2012-08-15 | 中国人民解放军国防科学技术大学 | Mass storage system monitoring method integrating heterogeneous storage devices |
CN103123602A (en) * | 2011-11-18 | 2013-05-29 | 阿里巴巴集团控股有限公司 | Abnormal alarming monitoring method based on java and device thereof |
CN105930357A (en) * | 2016-04-07 | 2016-09-07 | 深圳市慧动创想科技有限公司 | Distributed file system, and data node data storage processing method and device |
CN106484587A (en) * | 2015-08-26 | 2017-03-08 | 华为技术有限公司 | A kind of NameSpace management method, device and computer system |
CN109196474A (en) * | 2016-05-17 | 2019-01-11 | 微软技术许可有限责任公司 | Distributed operation control in computing system |
CN110109767A (en) * | 2019-05-10 | 2019-08-09 | 重庆八戒电子商务有限公司 | A kind of analysis method and system of distributed system availability |
CN110413437A (en) * | 2019-07-26 | 2019-11-05 | 济南浪潮数据技术有限公司 | Network namespace abnormality eliminating method, device, equipment and readable storage medium storing program for executing |
CN110908611A (en) * | 2019-11-24 | 2020-03-24 | 浪潮电子信息产业股份有限公司 | Block service starting method, device, equipment and medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2041642C (en) * | 1989-09-14 | 1997-03-04 | Tadamitsu Ryu | Temporary center system in distributed data base system |
CN101114867A (en) * | 2006-07-25 | 2008-01-30 | 深圳Tcl工业研究院有限公司 | Multi-channel synchronization transmitting method and system |
CN101621405A (en) * | 2009-07-07 | 2010-01-06 | 中兴通讯股份有限公司 | Distributed type management monitoring system, monitoring method and establishing method thereof |
US7809828B2 (en) * | 2007-04-11 | 2010-10-05 | International Business Machines Corporation | Method for maintaining state consistency among multiple state-driven file system entities when entities become disconnected |
-
2011
- 2011-03-15 CN CN 201110061037 patent/CN102110041B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2041642C (en) * | 1989-09-14 | 1997-03-04 | Tadamitsu Ryu | Temporary center system in distributed data base system |
CN101114867A (en) * | 2006-07-25 | 2008-01-30 | 深圳Tcl工业研究院有限公司 | Multi-channel synchronization transmitting method and system |
US7809828B2 (en) * | 2007-04-11 | 2010-10-05 | International Business Machines Corporation | Method for maintaining state consistency among multiple state-driven file system entities when entities become disconnected |
CN101621405A (en) * | 2009-07-07 | 2010-01-06 | 中兴通讯股份有限公司 | Distributed type management monitoring system, monitoring method and establishing method thereof |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103123602A (en) * | 2011-11-18 | 2013-05-29 | 阿里巴巴集团控股有限公司 | Abnormal alarming monitoring method based on java and device thereof |
CN103123602B (en) * | 2011-11-18 | 2016-04-27 | 阿里巴巴集团控股有限公司 | Based on abnormal alarm method for supervising and the device thereof of java |
CN102638378B (en) * | 2012-02-22 | 2014-05-21 | 中国人民解放军国防科学技术大学 | Mass storage system monitoring method integrating heterogeneous storage devices |
CN102638378A (en) * | 2012-02-22 | 2012-08-15 | 中国人民解放军国防科学技术大学 | Mass storage system monitoring method integrating heterogeneous storage devices |
CN106484587A (en) * | 2015-08-26 | 2017-03-08 | 华为技术有限公司 | A kind of NameSpace management method, device and computer system |
CN106484587B (en) * | 2015-08-26 | 2019-07-19 | 华为技术有限公司 | A kind of NameSpace management method, device and computer system |
CN105930357B (en) * | 2016-04-07 | 2019-12-27 | 深圳市慧动创想科技有限公司 | Distributed file system and data node data storage processing method and device |
CN105930357A (en) * | 2016-04-07 | 2016-09-07 | 深圳市慧动创想科技有限公司 | Distributed file system, and data node data storage processing method and device |
CN109196474A (en) * | 2016-05-17 | 2019-01-11 | 微软技术许可有限责任公司 | Distributed operation control in computing system |
CN109196474B (en) * | 2016-05-17 | 2021-12-03 | 微软技术许可有限责任公司 | Distributed operation control in a computing system |
CN110109767A (en) * | 2019-05-10 | 2019-08-09 | 重庆八戒电子商务有限公司 | A kind of analysis method and system of distributed system availability |
CN110413437A (en) * | 2019-07-26 | 2019-11-05 | 济南浪潮数据技术有限公司 | Network namespace abnormality eliminating method, device, equipment and readable storage medium storing program for executing |
CN110413437B (en) * | 2019-07-26 | 2022-06-17 | 济南浪潮数据技术有限公司 | Network namespace exception handling method, device and equipment and readable storage medium |
CN110908611A (en) * | 2019-11-24 | 2020-03-24 | 浪潮电子信息产业股份有限公司 | Block service starting method, device, equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
CN102110041B (en) | 2013-07-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102110041B (en) | System and method for monitoring storage space of distributed file system | |
US20230342271A1 (en) | Performance-Based Prioritization For Storage Systems Replicating A Dataset | |
US10733053B1 (en) | Disaster recovery for high-bandwidth distributed archives | |
US11934260B2 (en) | Problem signature-based corrective measure deployment | |
US11941288B1 (en) | Servicing write operations in a cloud-based storage system | |
US10552046B2 (en) | Automatic tiering of storage using dynamic grouping | |
US11966841B2 (en) | Search acceleration for artificial intelligence | |
US8244671B2 (en) | Replica placement and repair strategies in multinode storage systems | |
JP4896593B2 (en) | Performance monitoring method, computer and computer system | |
WO2019178470A1 (en) | Servicing i/o operations in a cloud-based storage system | |
US20200117534A1 (en) | Online failure span determination | |
US11210133B1 (en) | Workload mobility between disparate execution environments | |
US11604583B2 (en) | Policy based data tiering | |
CN105487823B (en) | A kind of method and device of Data Migration | |
CN110086888B (en) | Multi-cluster dynamic load method and device based on RabbitMQ and electronic equipment | |
US11847025B2 (en) | Storage system parity based on system characteristics | |
CN103946846A (en) | Use of virtual drive as hot spare for RAID group | |
US11494692B1 (en) | Hyperscale artificial intelligence and machine learning infrastructure | |
US20100268689A1 (en) | Providing information relating to usage of a simulated snapshot | |
CN105630418A (en) | Data storage method and device | |
CN103763383A (en) | Integrated cloud storage system and storage method thereof | |
CN103761059A (en) | Multi-disk storage method and system for mass data management | |
WO2016180049A1 (en) | Storage management method and distributed file system | |
US11797211B2 (en) | Expanding data structures in a storage system | |
Li et al. | Procode: A proactive erasure coding scheme for cloud storage systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20201116 Address after: 215100 No. 1 Guanpu Road, Guoxiang Street, Wuzhong Economic Development Zone, Suzhou City, Jiangsu Province Patentee after: SUZHOU LANGCHAO INTELLIGENT TECHNOLOGY Co.,Ltd. Address before: 100085 Beijing, Haidian District on the road to information on the ground floor, building 2-1, No. 1, C Patentee before: Inspur (Beijing) Electronic Information Industry Co.,Ltd. |