CN102541696A - Fault clearing method and device for comparison system - Google Patents
Fault clearing method and device for comparison system Download PDFInfo
- Publication number
- CN102541696A CN102541696A CN2010106124777A CN201010612477A CN102541696A CN 102541696 A CN102541696 A CN 102541696A CN 2010106124777 A CN2010106124777 A CN 2010106124777A CN 201010612477 A CN201010612477 A CN 201010612477A CN 102541696 A CN102541696 A CN 102541696A
- Authority
- CN
- China
- Prior art keywords
- comparison
- server
- characteristic
- fault
- break down
- 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
Images
Landscapes
- Hardware Redundancy (AREA)
Abstract
The invention discloses a fault clearing method and a fault clearing device for a comparison system. The comparison system comprises a plurality of comparison servers which store feature data respectively and compares the stored feature data. The fault clearing method comprises: monitoring the comparison servers; and when monitoring one or a plurality of comparison servers have a fault, driving the normal comparison servers to load the feature data stored in the faulted comparison server at the same time. The fault clearing method and the fault clearing device can clear faults in the comparison system in an automatic, quick and economic manner, thereby restoring the normal operation of the comparison system.
Description
Technical field
The present invention relates to aspect ratio to technical field, particularly a kind of fault handling method and device, comparison system that compares in the system.
Background technology
At present extensive comparison system and ultra-large comparison system; Like the fingerprint characteristic comparison in the comparison of the face characteristic in the Identification of Images system, the system of fingerprints; The system data amount generally can reach millions of even tens million of, more than one hundred million; Every characteristic is generally being counted KB between tens of KB, and the characteristic total amount in the comparison system can reach several GB to hundreds of GB.
Generally speaking, individual server can't be accomplished the comparison task of data volume like this, adopts a plurality of comparison servers usually, forms distributed comparison system.In distributed comparison system, when expendable soft, hardware fault appearred in one or more comparison servers, the time that the needs cost is long was carried out system recovery work, causes service disconnection, influences the normal use of system.
If use the mode of redundancy backup; Promptly each comparison server is equipped with backup comparison server; If the comparison server breaks down, then automatically switch to said backup comparison server, this mode need expend more hardware resource; And need corresponding software support, thereby significantly increase system cost.
But correlation technique also application reference number is 200810111412.7 one Chinese patent application, and this patent discloses a kind of system and method that is used for fingerprint comparison.
Summary of the invention
The problem that the present invention will solve is when the comparison system breaks down, be difficult to economical and efficiently mode fault is recovered to handle.
For addressing the above problem, the present invention provides a kind of fault handling method of comparing in the system, and said comparison system comprises a plurality of comparison servers, stores characteristic respectively, and based on the characteristic that the stores processing of comparing, said fault handling method comprises:
Monitor said a plurality of comparison server;
When monitoring one or more comparison servers when breaking down, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
Optional, also be included in said respectively comparing characteristic that server loads the comparison server stores that breaks down and accomplishing of not breaking down after, create the index of characteristic in each comparison server that does not break down.
Optional, the said a plurality of comparison servers of said monitoring comprise: regularly inquire about the running status of each comparison server, and judge whether said comparison server breaks down.
Optional, drive the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and comprise: drive said do not break down respectively compare the characteristic that server on average loads the said comparison server stores that breaks down.
Optional, drive the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and comprise: drives said do not break down respectively compare the characteristic of server according to the available separately said comparison server stores that breaks down of storage space situation loading.
Optional; Also comprise: during the said comparison server that does not break down loads said characteristic; If receive the comparison task that need compare with the characteristic of the said comparison server stores that breaks down, then the said comparison task of buffer memory is accomplished until the said comparison server that does not break down and is loaded said characteristic.
Optional, also comprise: when one or more comparison servers break down in monitoring said comparison system, send warning.
Optional, said characteristic behaviour face characteristic or fingerprint characteristic data.
For addressing the above problem; The present invention also provides a kind of fault treating apparatus of comparing in the system; Said comparison system comprises a plurality of comparison servers, stores characteristic respectively, and based on the characteristic that the stores processing of comparing; Said fault treating apparatus comprises main control server, and said main control server comprises:
The failure monitoring unit is used to monitor said a plurality of comparison server;
Fault processing unit, be used for when said failure monitoring unit monitors when one or more comparison servers break down, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
Optional, also comprise the index creation unit, be used for after said respectively comparing characteristic that server loads the comparison server stores that breaks down and accomplishing of not breaking down, creating the index of characteristic in each comparison server that does not break down.
Optional, the said a plurality of comparison servers of said failure monitoring unit monitors are meant: the running status of each comparison server is regularly inquired about in said failure monitoring unit, and judges whether said comparison server breaks down.
Optional, said fault processing unit drives the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and is meant: said fault processing unit drive said do not break down respectively compare the characteristic that server on average loads the said comparison server stores that breaks down.
Optional, said fault processing unit drives the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and is meant: said fault processing unit drive said do not break down respectively compare server loads the said comparison server stores that breaks down according to available separately storage space situation characteristic.
Optional; Also comprise buffer unit; During the said comparison server that does not break down loads said characteristic; If receive the comparison task of comparing with the characteristic of the said comparison server stores that breaks down, then the said comparison task of buffer memory is accomplished until the comparison server that does not break down and is loaded said characteristic.
Optional, also comprise backup master server, be used for when said main control server breaks down, switching to main control server.
Optional, also comprise alarm unit, be used for when the one or more comparison servers of said comparison system break down, sending warning when said failure monitoring unit monitors.
Optional, said characteristic behaviour face characteristic or fingerprint characteristic data.
For addressing the above problem; The present invention also provides a kind of comparison system, comprises allocation units and a plurality of comparison server, and said a plurality of comparison servers store characteristic respectively; And based on the characteristic that the stores processing of comparing; Said allocation units obtain the comparison task, and distribute said comparison task to corresponding comparison server based on the characteristic of each comparison server stores, also comprise above-mentioned fault treating apparatus.
Optional, said comparison system is a distributed comparison system.
Optional, said comparison behaviour face comparison system of system or fingerprint comparison system.
Optional, said allocation units are arranged in the said main control server.
Compared with prior art, the present invention has the following advantages:
Through all the comparison servers in the monitoring comparison system; When wherein one or more comparison servers break down; Drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server; Make other comparison servers that do not break down to bear the comparison work of the comparison server that breaks down, thereby need not under the situation of manual intervention, can make said comparison system accomplish Fault Processing in the short period of time automatically.
Be equipped with backup comparison server owing to need not compare server, therefore do not need bigger hardware and software input, reduced cost for each.
Description of drawings
Fig. 1 is the schematic flow sheet of the fault handling method in the comparison provided by the invention system;
Fig. 2 is a kind of example structure synoptic diagram of the fault treating apparatus in the comparison provided by the invention system;
Fig. 3 is the another kind of example structure synoptic diagram of the fault treating apparatus in the comparison provided by the invention system;
Fig. 4 is a comparison system implementation illustration intention provided by the invention.
Embodiment
For make above-mentioned purpose of the present invention, feature and advantage can be more obviously understandable, does detailed explanation below in conjunction with the accompanying drawing specific embodiments of the invention.
Set forth detail in the following description so that make much of the present invention.But the present invention can be different from alternate manner described here and implements with multiple, and those skilled in the art can do similar popularization under the situation of intension of the present invention.Therefore the present invention does not receive the restriction of following disclosed embodiment.
Described in background technology, when expendable soft, hardware fault appears in the one or more comparison servers in the comparison system, be difficult to economical and efficiently mode fault is handled to recover the normal use of said comparison system as early as possible.For this reason, the invention provides a kind of fault handling method of comparing in the system.
Fig. 1 is the schematic flow sheet of the fault handling method in the comparison provided by the invention system.As shown in Figure 1, said fault handling method comprises:
Step S101, the included a plurality of comparison servers of monitoring comparison system;
Step S102, when monitoring one or more comparison servers when breaking down, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
For implementing above-mentioned fault handling method, the invention provides a kind of fault treating apparatus of comparing in the system.Fig. 2 is a kind of example structure synoptic diagram of the fault treating apparatus in the comparison provided by the invention system; As shown in Figure 2; Said comparison system comprises a plurality of comparison servers; Be respectively comparison server 1, comparison server 2, comparison server 3...... comparison server n, store characteristic respectively in these comparison servers, and based on the characteristic that the stores processing of comparing.Said fault treating apparatus comprises main control server 10, and said main control server 10 comprises: failure monitoring unit 101 is used to monitor said a plurality of comparison server; Fault processing unit 102 is used for monitoring one or more comparison servers when breaking down when said failure monitoring unit 101, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
During practical implementation, said comparison system is an extensive or ultra-large distributed comparison system, for example people's face comparison system or fingerprint comparison system, this moment then said characteristic behaviour face characteristic or fingerprint characteristic data.Each comparison server stored characteristic is to supply comparison to handle in the local disk separately through from the database of said comparison system, being carried in; Situation according to each comparison server performance and storage space; The amount of the characteristic that is loaded is not necessarily identical; And can there be identical part in the characteristic that is loaded, and (characteristic that is same content can be loaded on a plurality of comparison servers, when receiving the comparison task that need compare with this characteristic; Can distribute the processing of comparing of which comparison server by said main control server 10), also can there be different portions.
Monitoring said a plurality of comparison servers in said failure monitoring unit 101 specifically are the running statuses of taking regularly to inquire about each comparison server, comprise that inquiry respectively compares the various performance datas of server, and judge with this whether said comparison server breaks down.Said fault had both comprised the hardware fault (for example local disk damage) that said comparison server occurs, and also comprised the software fault (for example comparison program is made mistakes) of comparing server.In case monitor one or more comparison servers when breaking down; Then said comparison system is in abnormal operating state; Because if comparison task need compare with the characteristic of the said comparison server stores that breaks down (suppose other comparison servers all do not load this characteristic) is arranged this moment; Then owing to the fault of this comparison server can not be accomplished said comparison system should the comparison task, cause service disconnection, influence the normal use of comparison system.At this time, if the comparison server that grade breaks down is fixed a breakdown back and reloaded characteristic, will spend the long time; If use the mode of redundancy backup, promptly each comparison server is equipped with backup comparison server, if the comparison server breaks down; Then automatically switch to said backup comparison server, this mode then need expend more hardware resource, and needs corresponding software support; In addition, because the comparison server stores has the number of characteristics data, and accomplish the comparison task that is assigned with based on these characteristics; Therefore higher to the system resource and the performance requirement of comparison server; Likewise, backup comparison server is also had equal high requirement, thereby significantly increase the cost of comparison system.And the present invention monitors one or more comparison servers when breaking down through said failure monitoring unit 101; Trigger said fault processing unit 102 drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server, thereby can make said comparison system recovery operate as normal automatically, quickly.
Particularly, said fault processing unit 102 drives said respectively comparing the common characteristic that loads the comparison server stores that breaks down of server and can taking following dual mode at least of not breaking down:
Mode one, said fault processing unit 102 drive said do not break down respectively compare the characteristic that server on average loads the said comparison server stores that breaks down.Consult Fig. 2; Suppose that in the comparison system performance of comparison server 1, comparison server 2, comparison server 3...... comparison server n is basic identical, and enough free memories are all arranged; The amount of the characteristic of each comparison server stores is also identical, for example is the data volume of S.Fault has taken place in server 3 if said failure monitoring unit 101 monitors comparison; Then trigger said fault processing unit 102 and drive the characteristic that n-1 the comparison server that does not break down on average loads the said comparison server stores that breaks down; The data volume that is each comparison said characteristic that server loaded is S/ (n-1), will accelerate to compare system failure processing speed greatly like this.Because the fault handling time of comparison system, only the loading velocity with the stored characteristic of the comparison server that breaks down is relevant.Suppose: 5,000 ten thousand characteristics are arranged, every data 6KB, 300GB data volume altogether; Divide 10 comparison server process, then each characteristic of on average being responsible for is 30GB, when one of them comparison server breaks down; Other 9 comparison servers load the stored characteristic (30GB) of comparison server that breaks down, and then each on average is responsible for loading about 3.4GB, loads the 10MB data computation with per second; Then need 340 seconds; Add index reconstruction after the loading etc., need 20 seconds approximately, promptly can accomplish in about 6 minutes altogether Fault Processing.Said completion only is to instigate the comparison system can operate as normal to Fault Processing, accomplishes the comparison task, as for the comparison server that specifically breaks down, need wait for that also person skilled solves.
Need to prove; In aforesaid way; It is enough only to need to satisfy the free memory of supposing of respectively comparing server; Can load the stored characteristic of the comparison server that breaks down and get final product through the data volume after average, whether identical, stored whether identical these hypothesis have been merely this mode of being more readily understood as for each data volume of comparing the performance of server characteristic, are not necessary condition.
Mode two, said fault processing unit 102 drive said do not break down respectively compare server loads the said comparison server stores that breaks down according to available separately storage space situation characteristic.In first kind of mode; The free memory of having supposed respectively to compare server is enough; But in actual implementation process; Because the data volume of characteristic is bigger, performance and available storage area different that each compares server need said fault processing unit 102 drive the characteristic that corresponding comparison server loads the said comparison server stores that breaks down according to available storage space situation of each comparison server.Under this mode; Need the available storage space advisory said fault processing unit 102 of said failure monitoring unit 101, carry out handled according to the free memory situation of difference comparison server by said fault processing unit 102 with each comparison server.This mode has applicability widely; When reality is implemented; Can analyze the free memory situation of each comparison server earlier,, then can take mode one if the free memory of each comparison server is enough; If the free memory of certain comparison server is not enough, then take to make this comparison server to load an amount of characteristic even can not load characteristic fully.
Certainly, can also take other modes, such as letting the strong comparison server of performance add to carry some characteristics, and let the comparison server of poor-performing load some characteristics less even can not load characteristic.
On the basis of the foregoing description, can also take another kind of embodiment.Fig. 3 is the another kind of example structure synoptic diagram of the fault treating apparatus in the comparison provided by the invention system.As shown in Figure 3; Said fault treating apparatus also comprises index creation unit 104; Be used for after said respectively comparing characteristic that server loads the comparison server stores break down and accomplishing of not breaking down, creating the index of characteristic in each comparison server that does not break down.Particularly; Said index creation unit 104 can be arranged among the said main control server 10; Break down at certain or certain several comparison servers; And after accomplishing by other respectively comparing of not breaking down characteristic that server loads the comparison server stores break down; Because change (having increased characteristic) has taken place stored characteristic in the said comparison server that does not break down, so need rebulid the index to characteristic in these comparison servers by said index creation unit 104, said like this main control server 10 just can be according to the corresponding comparison server of setting up of the new index assignment processing of comparing.
Said fault treating apparatus also comprises alarm unit 103, is used for sending warning when the one or more comparison servers of said comparison system break down when said failure monitoring unit 101 monitors.Particularly; In the present embodiment, as shown in Figure 3, said alarm unit 103 can be arranged among the said main control server 10; The warning that said alarm unit 103 sends can be reminded out of order generation, and the notice person skilled is recovered the comparison server that breaks down.
Said fault treating apparatus also comprises buffer unit 105; Be used for during the said comparison server that does not break down loads said characteristic; If receive the comparison task of comparing with the characteristic of the said comparison server stores that breaks down, then the said comparison task of buffer memory is accomplished until the comparison server that does not break down and is loaded said characteristic.Particularly, as shown in Figure 3 in the present embodiment, said buffer unit 105 is arranged among the said main control server 10.Because when reality is implemented; After one or more comparison servers break down; Still can receive a large amount of comparison tasks; If just in time receive the comparison task that to compare with the characteristic of the said comparison server stores that breaks down this moment, the comparison processing to this comparison task, the said comparison server that does not break down can't be provided so, therefore during loading said characteristic; Need the buffer memory should the comparison task; Accomplish to load said characteristic until the comparison server that do not break down, and after rebuliding index by said index creation unit 104, said main control server 10 just can be according to the corresponding comparison server of the new index assignment of setting up to the processing of comparing of said comparison task.
In addition, said fault treating apparatus also can comprise backup master server 10 ', is used for when said main control server 10 breaks down, switching to main control server.Said backup master server 10 ' can have identical structure and bear identical functions with said main control server 10; As shown in Figure 3; Said backup master server 10 ' can comprise failure monitoring unit 101 ', fault processing unit 102 ', alarm unit 103 ', index creation unit 104 ', buffer unit 105 ', the annexation of each unit and effect and said main control server 10 basically identicals.Difference is to some extent; When said main control server 10 operate as normal; Monitor each other between said main control server 10 and the said backup master server 10 ', that is: by said failure monitoring unit 101 ' running status of said main control server 10 is monitored, monitor by the 101 pairs of said backup master servers 10 ' in said failure monitoring unit; If said failure monitoring unit 101 ' monitors said main control server 10 and breaks down; Then said backup master server 10 ' switches to main control server, bears all working of said main control server 10 before breaking down, and sends warning by said alarm unit 103 '; If said failure monitoring unit 101 monitors said backup master server 10 ' and breaks down, then send warning by said alarm unit 103.In said main control server 10 normal work period, specifically not working in each unit of other of said backup master server 10 ', only is in holding state.Can be between said main control server 10 and the said backup master server 10 ' through the network interconnection (for example Fast Ethernet); And; Because what said main control server 10 and said backup master server 10 ' were realized act as the monitoring of comparison server and mutual monitoring; Needed system resource requires lower, so can use common PC or the lower server of configuration to realize.
Based on above-mentioned fault treating apparatus, the invention provides a kind of comparison system.Fig. 4 is a comparison system implementation illustration intention provided by the invention.As shown in Figure 4; Said comparison system comprises allocation units 106 and a plurality of comparison servers; Said a plurality of comparison server stores characteristic respectively; And based on the characteristic that the stores processing of comparing, said characteristic loads through database 20, and said allocation units 106 obtain the comparison task; And distributing the processing of comparing of said comparison task to corresponding comparison server based on the characteristic of each comparison server stores, said comparison system also comprises above-mentioned fault treating apparatus.In the present embodiment; Said allocation units 106 are arranged in the main control server 10 of said fault treating apparatus (also can be provided with same allocation units in said backup master server 10 '; The concrete structure of said backup master server 10 ' can not specifically illustrate structure with reference to figure 3 and main control server 10 shown in Figure 4 in Fig. 4).During practical implementation, said comparison system is an extensive or ultra-large distributed comparison system, for example people's face comparison system or fingerprint comparison system, this moment then said characteristic behaviour face characteristic or fingerprint characteristic data.
Consult Fig. 4; When the user need compare; Client through said comparison system is initiated the comparison task; Said comparison system can comprise a plurality of clients, and like client 1, client 2, client 3...... client m, these clients are transmitted the comparison tasks through network to said main control server 10; Distribute processings of comparing of said comparison task to corresponding comparison server by the allocation units 106 of said main control server 10 based on the characteristic of each comparison server stores, will compare the client that result after the processing feeds back to the said comparison task of initiation again.Said comparison system comprises fault treating apparatus according to the invention, and the practical implementation process of said fault treating apparatus can be carried out with reference to the embodiment of above-mentioned fault treating apparatus.The said main control server 10 of said fault treating apparatus can be interconnected through network and each comparison server; When the said failure monitoring unit 101 in the said main control server 10 monitors one or more comparison servers when breaking down; Trigger said fault processing unit 102 drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server; Again by said index creation unit 104 after said respectively comparing characteristic that server loads the comparison server stores break down and accomplishing of not breaking down, create the index of characteristic in each comparison server that does not break down.After the establishment of accomplishing new index, said allocation units 106 are based on the index of new establishment, and distribution ratio is to the processing of comparing of task to corresponding comparison server.In addition; If during the said comparison server that does not break down loads said characteristic; If receive the comparison task of comparing with the characteristic of the said comparison server stores that breaks down; Then accomplished until the comparison server that do not break down by the said comparison task of said buffer unit 105 buffer memorys and load said characteristic, said allocation units 106 obtain the said comparison task of 105 buffer memorys of said buffer unit, distribute to the processing of comparing of corresponding comparison server.Need to prove, because the said comparison task of institute's buffer memory just receives during fault handling, so after said comparison system recovery operate as normal, said comparison task that should 105 buffer memorys of the said buffer unit of priority processing.
Other practical implementations of fault handling method in said comparison system and the said comparison system can be carried out with reference to the fault treating apparatus in the said comparison system, repeat no more at this.
To sum up, fault handling method in the comparison provided by the invention system and device, comparison system have following beneficial effect at least:
Through all the comparison servers in the monitoring comparison system (especially extensive or ultra-large distributed comparison system); When wherein one or more comparison servers break down; Drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server; Make other comparison servers that do not break down to bear the comparison work of the comparison server that breaks down; Thereby need not under the situation of manual intervention, can make said comparison system accomplish Fault Processing in the short period of time automatically.
Be equipped with backup comparison server owing to need not compare server, therefore do not need bigger hardware and software input, reduced cost for each.
Though the present invention with preferred embodiment openly as above; But it is not to be used for limiting the present invention; Any those skilled in the art are not breaking away from the spirit and scope of the present invention; Can utilize the method and the technology contents of above-mentioned announcement that technical scheme of the present invention is made possible change and modification, therefore, every content that does not break away from technical scheme of the present invention; To any simple modification, equivalent variations and modification that above embodiment did, all belong to the protection domain of technical scheme of the present invention according to technical spirit of the present invention.
Claims (21)
1. fault handling method of comparing in the system, said comparison system comprises a plurality of comparison servers, stores characteristic respectively, and based on the characteristic that the stores processing of comparing, it is characterized in that, comprising:
Monitor said a plurality of comparison server;
When monitoring one or more comparison servers when breaking down, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
2. fault handling method according to claim 1; It is characterized in that; After also being included in said respectively comparing characteristic that server loads the comparison server stores break down and accomplishing of not breaking down, create the index of characteristic in each comparison server that does not break down.
3. fault handling method according to claim 1 is characterized in that, the said a plurality of comparison servers of said monitoring comprise: regularly inquire about the running status of each comparison server, and judge whether said comparison server breaks down.
4. fault handling method according to claim 1; It is characterized in that, drive the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and comprise: drive said do not break down respectively compare the characteristic that server on average loads the said comparison server stores that breaks down.
5. fault handling method according to claim 1; It is characterized in that, drive the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and comprise: drives said do not break down respectively compare the characteristic of server according to the available separately said comparison server stores that breaks down of storage space situation loading.
6. fault handling method according to claim 1; It is characterized in that; Also comprise: during the said comparison server that does not break down loads said characteristic; If receive the comparison task that need compare with the characteristic of the said comparison server stores that breaks down, then the said comparison task of buffer memory is accomplished until the said comparison server that does not break down and is loaded said characteristic.
7. fault handling method according to claim 1 is characterized in that, also comprises: when one or more comparison servers break down in monitoring said comparison system, send warning.
8. fault handling method according to claim 1 is characterized in that, said characteristic behaviour face characteristic or fingerprint characteristic data.
9. fault treating apparatus of comparing in the system, said comparison system comprises a plurality of comparison servers, stores characteristic respectively; And based on the characteristic that the stores processing of comparing; It is characterized in that, comprise main control server, said main control server comprises:
The failure monitoring unit is used to monitor said a plurality of comparison server;
Fault processing unit, be used for when said failure monitoring unit monitors when one or more comparison servers break down, drive do not break down respectively compare the common characteristic that loads the comparison server stores that breaks down of server.
10. fault treating apparatus according to claim 9; It is characterized in that; Also comprise the index creation unit; Be used for after said respectively comparing characteristic that server loads the comparison server stores break down and accomplishing of not breaking down, creating the index of characteristic in each comparison server that does not break down.
11. fault treating apparatus according to claim 9; It is characterized in that; The said a plurality of comparison servers of said failure monitoring unit monitors are meant: the running status of each comparison server is regularly inquired about in said failure monitoring unit, and judges whether said comparison server breaks down.
12. fault treating apparatus according to claim 9; It is characterized in that said fault processing unit drives the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and is meant: said fault processing unit drive said do not break down respectively compare the characteristic that server on average loads the said comparison server stores that breaks down.
13. fault treating apparatus according to claim 9; It is characterized in that said fault processing unit drives the said common characteristic that loads the comparison server stores that breaks down of server of respectively comparing that does not break down and is meant: said fault processing unit drive said do not break down respectively compare server loads the said comparison server stores that breaks down according to available separately storage space situation characteristic.
14. fault treating apparatus according to claim 9; It is characterized in that; Also comprise buffer unit; During the said comparison server that does not break down loaded said characteristic, if receive the comparison task of comparing with the characteristic of the said comparison server stores that breaks down, then the said comparison task of buffer memory was accomplished until the comparison server that does not break down and is loaded said characteristic.
15. fault treating apparatus according to claim 9 is characterized in that, also comprises backup master server, is used for when said main control server breaks down, switching to main control server.
16. fault treating apparatus according to claim 9 is characterized in that, also comprises alarm unit, is used for when the one or more comparison servers of said comparison system break down, sending warning when said failure monitoring unit monitors.
17. fault treating apparatus according to claim 9 is characterized in that, said characteristic behaviour face characteristic or fingerprint characteristic data.
18. comparison system; Comprise allocation units and a plurality of comparison server, said a plurality of comparison servers store characteristic respectively, and based on the characteristic that the stores processing of comparing; Said allocation units obtain the comparison task; And distribute said comparison task to corresponding comparison server based on characteristic of each comparison server stores, and it is characterized in that, also comprise each described fault treating apparatus of claim 9～17.
19. comparison according to claim 18 system is characterized in that, said comparison system is a distributed comparison system.
20. comparison according to claim 18 system is characterized in that, said comparison behaviour face comparison system of system or fingerprint comparison system.
21. comparison according to claim 18 system is characterized in that said allocation units are arranged in the said main control server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106124777A CN102541696A (en) | 2010-12-29 | 2010-12-29 | Fault clearing method and device for comparison system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106124777A CN102541696A (en) | 2010-12-29 | 2010-12-29 | Fault clearing method and device for comparison system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102541696A true CN102541696A (en) | 2012-07-04 |
Family
ID=46348646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106124777A Pending CN102541696A (en) | 2010-12-29 | 2010-12-29 | Fault clearing method and device for comparison system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102541696A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201111084Y (en) * | 2007-07-27 | 2008-09-03 | 上海银晨智能识别科技有限公司 | Portrait contrasting system used for communal security system |
| CN101467132A (en) * | 2006-06-15 | 2009-06-24 | 摩托罗拉公司 | Method and system for distributing data processing units in a communication network |
| US20100031081A1 (en) * | 2004-03-31 | 2010-02-04 | Nec Corporation | Data Storage System and Control Method Thereof |
-
2010
- 2010-12-29 CN CN2010106124777A patent/CN102541696A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100031081A1 (en) * | 2004-03-31 | 2010-02-04 | Nec Corporation | Data Storage System and Control Method Thereof |
| CN101467132A (en) * | 2006-06-15 | 2009-06-24 | 摩托罗拉公司 | Method and system for distributing data processing units in a communication network |
| CN201111084Y (en) * | 2007-07-27 | 2008-09-03 | 上海银晨智能识别科技有限公司 | Portrait contrasting system used for communal security system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10152382B2 (en) | Method and system for monitoring virtual machine cluster | |
| CN1947096B (en) | Dynamic migration of virtual machine computer programs | |
| US9141491B2 (en) | Highly available server system based on cloud computing | |
| CN105095001B (en) | Virtual machine abnormal restoring method under distributed environment | |
| US10331472B2 (en) | Virtual machine service availability | |
| CN102622265A (en) | Method and system for task distribution | |
| CN103581225A (en) | Distributed system node processing task method | |
| CN104754048A (en) | Imitative organization structure of server cluster | |
| KR101586354B1 (en) | Communication failure recover method of parallel-connecte server system | |
| CN107544867B (en) | Method, device and system for recovering intelligent network service | |
| CN103441987A (en) | Method and device for managing dual-computer firewall system | |
| CN107844339B (en) | Task scheduling method, plug-in and server | |
| CN202870563U (en) | Distributed comprehensive monitoring system | |
| CN111988347B (en) | Data processing method of board hopping machine system and board hopping machine system | |
| US20180336061A1 (en) | Storing file portions in data storage space available to service processors across a plurality of endpoint devices | |
| CN105591964A (en) | Device and method for overload protection for Internet system | |
| CN108445857A (en) | A kind of 1+N redundancy scheme design methods of SCADA system | |
| CN101155071B (en) | Method, device and communication system for selecting local domain supervisor | |
| CN102541696A (en) | Fault clearing method and device for comparison system | |
| CN105897487B (en) | Equipment management method and device for operation and maintenance system | |
| CN115470009B (en) | Memory resource management method, system, device and medium | |
| CN103139066B (en) | A kind of changing method of resource, main control device and node device | |
| CN112269693B (en) | Node self-coordination method, device and computer readable storage medium | |
| JP2009123238A (en) | Controller, computer system, process control method of controller, computer control method of computer system, computer control program and process control program | |
| US11210187B1 (en) | Computer cluster with adaptive quorum rules |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120704 |