CN1652616B - Adaptive system managing method - Google Patents
Adaptive system managing method Download PDFInfo
- Publication number
- CN1652616B CN1652616B CN 200410004120 CN200410004120A CN1652616B CN 1652616 B CN1652616 B CN 1652616B CN 200410004120 CN200410004120 CN 200410004120 CN 200410004120 A CN200410004120 A CN 200410004120A CN 1652616 B CN1652616 B CN 1652616B
- Authority
- CN
- China
- Prior art keywords
- storage management
- internal memory
- module
- pointer
- management
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Debugging And Monitoring (AREA)
- Exchange Systems With Centralized Control (AREA)
Abstract
The method includes following steps: selecting first management program from multiple management programs in the same class provided by system and running the first management program; recording operational aspect of each module in system when first management program is in running; based on recorded operational aspect of each module in system, selecting second management program suitable to running environment for each module in multiple management programs; running the said second management program to carry out system management.
Description
Technical field
The present invention relates to the management method in the communication system, particularly a kind of adaptive method for managing system.
Background technology
In the telesoftware system, require system operation controlled, real-time, this just need manage system, and such as must be controlled to the time of memory management (application and release), guaranteeing also can not become under worst condition, it be one uncontrollable to accept.
In order to guarantee the controlled of total system, generally be that the system management algorithm is improved, use optimum algorithm to come guaranteed efficiency or utilization factor.But because the complexity of system, the ruuning situation of system varies, and does not have a kind of algorithm can guarantee that system just necessarily is in optimum condition in each operation occasion usually.
In order to address the above problem, a kind of method is to provide an algorithm all more common under each environment according to estimating, allows each module of system each time of running of unified the use.Though this method is simple, but system operation is complicated and changeable, is difficult to find out the algorithm that all suits when each module respectively moved, and also is difficult to estimate the ruuning situation of system's reality, thereby cause in some cases that the worst condition of system becomes unacceptable, even cause system crash.
Another balance method is that multiple management algorithm is provided in system, and the system that stipulates in advance uses certain management algorithm.But because the supplier of algorithm separates usually with the user, cause the user not know the adaptation occasion of algorithm, might misapply and cause effect poorer, more likely be because algorithm user self do not know yet system practical operation situation be meet which kind of algorithm model and can't effectively make one's options, cause this scheme in fact also to be difficult to use.Simultaneously, link up, train or, all can bring very big workload and increase cost no matter be modification of program.
Summary of the invention
The object of the present invention is to provide a kind of adaptive method for managing system, to solve the problem that the method for managing system selected in the prior art can not the adaptive system practical operation situation.
The invention provides following technical proposals:
A kind of adaptive method for managing system, the method comprising the steps of:
A, selection first storage management and operation from a plurality of storage managements that system provides;
System uses each module of internal memory to apply for internal memory at every turn when B, record operation first storage management size and release conditions;
C, according to the internal memory application of each module of system of record with discharge ruuning situation, from described multiple storage management, select to be fit to second storage management of running environment for each module of using internal memory;
D, described second storage management of operation carry out memory management.
According to said method:
The common memory management program that described first storage management provides for system.
Described first storage management is selected from a plurality of storage managements at random.
Each module of system no longer writes down ruuning situation when moving second storage management.
Stop the ruuning situation of register system when bringing into operation second storage management, and the operation some cycles after with second storage management as first storage management, continue step B again.
The pointer that points to the record function and first storage management respectively is set in the internal memory application interface that system provides, the pointer that points to the record function and first storage management respectively is set in the internal memory realizing interface that system provides, the pointer that points to the record function in internal memory application interface and internal memory realizing interface is called, and writes down the internal memory application in service of each module and the situation that internal memory discharges by this record function.
When operation second storage management, the pointer that points to the record function in internal memory application interface and the internal memory realizing interface is put sky, the pointer that points to first storage management is set to point to second storage management.
The pointer that points to the record function is set respectively in internal memory application interface that system provides and internal memory realizing interface, when invoke memory application interface and internal memory realizing interface move first storage management, this record function is called earlier, and by this record function call first storage management; When operation second storage management, described pointer is directly pointed to this second storage management.
The present invention is by writing down ruuning situation automatically, and according to the suitable system supervisor of ruuning situation selection, make the supervisory routine maker can concentrate one's energy to consider the business of oneself, and needn't consider how to select supervisory routine, thereby can put forward work efficiency, also can the assurance system move more scientific and reasonable simultaneously.
Description of drawings
Fig. 1 is a switch system memory management synoptic diagram;
Fig. 2 is the process flow diagram of the embodiment of the invention.
Embodiment
In the present invention, for system provides a plurality of similar supervisory routines, at first from a plurality of supervisory routines, select first supervisory routine to move, and call the ruuning situation that logging program writes down system when moving first supervisory routine; Behind first supervisory routine operation some cycles,, from described multiple supervisory routine, select to be fit to second supervisory routine operation of system running environment, by moving the second supervisory routine management system according to the ruuning situation of record.
In order to guarantee when calling first supervisory routine, in time to call logging program, can realize by following dual mode:
1, the pointer that points to the logging program and first supervisory routine respectively is set in the supervisor call interface that system provides, when the supervisor call interface moves first supervisory routine logging program is called when calling, writes down the ruuning situation of each module by this logging program.When operation second supervisory routine, the pointer that points to first supervisory routine is set to point to this second supervisory routine; If do not need to write down the ruuning situation of each module this moment, then will point to the pointer of logging program and put empty getting final product.
2, the pointer that points to logging program only is set in the supervisor call interface that system provides, when calling the supervisor call interface and move first supervisory routine, this logging program is called earlier, and calls first supervisory routine by this logging program; When operation second supervisory routine, described pointer is directly pointed to this second supervisory routine.
Be that storage management is that the present invention is described in detail for example below with the supervisory routine:
Consult shown in Figure 1ly, in switch system, total system is divided into a lot of modules, such as communication module, and monitoring module etc.Each module all needs to use internal memory, and each module is also all inequality to the internal memory operating position, and the internal memory application of each module and release efficiency have been got conclusive effect to total system efficient.The distribution of internal memory and release etc. are finished by the storage management module.
To different internal memory operating positions, should use algorithms of different to carry out high-level efficiency and high usage that memory management could guarantee memory management.In order to guarantee that each module rationally uses internal memory, improve the reasonable use of the efficient of whole switch system and system to internal memory.The memory management of switch provides multiple algorithm, as the static memory algorithm, and piece internal memory algorithm, DOS internal memory algorithm, each algorithm are a storage management.Various algorithms all have the suitable applications occasion of oneself, are fit to the fixing dynamic application of each application length such as the piece internal memory, and efficiency ratio is higher, if but apply for that length in very large range occurs at random, then can cause serious internal memory waste.
Consult process flow diagram shown in Figure 2, the storage management of practical operation situation is selected to be fit to by system by following step:
Step 10: from a plurality of storage managements that system provides, select the operation of first storage management;
First storage management can be a general storage management, also can be an optional storage management.
Step 20: each module is applied for memory size, release conditions at every turn during by the operation of record function register system.
Provide internal memory application interface and internal memory realizing interface by system, two function pointer A, B are arranged respectively in internal memory application and realizing interface.Pointer A in the application interface is used to refer to the record function, and pointer B is used to refer to physical memory application algorithm routine; Pointer A in the realizing interface also is used to refer to the record function, and pointer B is used to refer to internal memory and discharges algorithm routine.Comprise the information that to apply for length and application module in the application interface, do not need selection algorithm; Also comprise the memory address pointer that will discharge in the realizing interface.
Make the pointer A in application interface and the realizing interface all point to the record function when initial, the internal memory application and the internal memory that make the pointer B that applies in interface and the realizing interface point to first storage management respectively discharge algorithm routine.
When internal memory application and internal memory realizing interface were called, pointer A was called, and just write down function and was called, and write down the situation that internal memory application in service and internal memory discharge by the record function; Simultaneously, pointer B also is called, and finishes physical memory application or memory discharge operation.
Step 30:,, judge the optimal algorithm that each module should be used under this running environment, i.e. second storage management according to the internal memory application and the release ruuning situation of each module of noting through operation after a while.
By the discriminant function that calling system provides, compare with the usable range of various algorithms according to internal memory application, internal memory release conditions and the current internal memory operating position of record function record, judge which kind of algorithm is each situation that records be fit to use respectively.
Step 40: move second storage management and carry out memory management.
After determining second internally stored program, the pointer B in application, the realizing interface is dynamically pointed to the rational management algorithm program of each module, simultaneously the A pointer in application and the realizing interface is put sky, record no longer when applying for release afterwards.
In step 20, the interface that provides also can only use a function pointer, and this pointer points to the record function at the beginning, and by the record function call first memory management algorithm routine.When rational second storage management of judging that each module should use, just this pointer is directly pointed to this program.In addition, also can not comprise module information in the interface, promptly interface only comprises needs application length information.In in interface, realizing, by getting task number (perhaps thread number) supervisor running environment information, automatic judge module information.
If ruuning situation is not to change but complicated dynamic change in the simple cycle, can be after a period of time at operation second storage management, pointer A is pointed to the record function again continue record ruuning situation, with second storage management as first storage management, ruuning situation according to record, reselect the appropriate algorithm that each module is used, i.e. second storage management by above-mentioned steps.
More than be that the present invention will be described for example with the storage management, but be not limited in this, other resource managements to system, as the route assignment algorithm, dynamically revise routing algorithm etc. according to network condition and all be suitable for, its principle is identical, promptly in system's actual motion, write down ruuning situation, and it is adaptive dynamically to carry out software flow according to practical operation situation.
Claims (7)
1. adaptive method for managing system is characterized in that comprising step:
A, selection first storage management and operation from a plurality of storage managements that system provides;
System uses each module of internal memory to apply for internal memory at every turn when B, record operation first storage management size and release conditions;
C, according to the internal memory application of each module of system of record with discharge ruuning situation, from described a plurality of storage managements, select to be fit to second storage management of running environment for each module of using internal memory;
D, described second storage management of operation carry out memory management.
2. the method for claim 1 is characterized in that, the common memory management program that described first storage management provides for system.
3. the method for claim 1 is characterized in that, described first storage management is selected from a plurality of storage managements at random.
4. the method for claim 1 is characterized in that, each module of system no longer writes down ruuning situation when moving second storage management.
5. the method for claim 1 is characterized in that, stops the ruuning situation of register system when bringing into operation second storage management, and the operation some cycles after with second storage management as first storage management, continue step B again.
6. the method for claim 1, it is characterized in that, the pointer that points to the record function and first storage management respectively is set in the internal memory application interface that system provides, the pointer that points to the record function and first storage management respectively is set in the internal memory realizing interface that system provides, the pointer that points to the record function in internal memory application interface and internal memory realizing interface is called, and writes down the internal memory application in service of each module and the situation that internal memory discharges by this record function.
7. method as claimed in claim 6, it is characterized in that, when operation second storage management, the pointer that points to the record function in internal memory application interface and the internal memory realizing interface is put sky, the pointer that points to first storage management is set to point to second storage management.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410004120 CN1652616B (en) | 2004-02-07 | 2004-02-07 | Adaptive system managing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410004120 CN1652616B (en) | 2004-02-07 | 2004-02-07 | Adaptive system managing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1652616A CN1652616A (en) | 2005-08-10 |
| CN1652616B true CN1652616B (en) | 2010-04-14 |
Family
ID=34867627
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410004120 Expired - Fee Related CN1652616B (en) | 2004-02-07 | 2004-02-07 | Adaptive system managing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1652616B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164290B (en) * | 2011-12-14 | 2017-08-04 | 腾讯科技（深圳）有限公司 | application memory management method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1393780A (en) * | 2001-06-28 | 2003-01-29 | 华为技术有限公司 | Adaptive dynamic memory management method |
| CN1427342A (en) * | 2001-12-21 | 2003-07-02 | 上海贝尔有限公司 | Internal storage management system and its distribution method |
-
2004
- 2004-02-07 CN CN 200410004120 patent/CN1652616B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1393780A (en) * | 2001-06-28 | 2003-01-29 | 华为技术有限公司 | Adaptive dynamic memory management method |
| CN1427342A (en) * | 2001-12-21 | 2003-07-02 | 上海贝尔有限公司 | Internal storage management system and its distribution method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1652616A (en) | 2005-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20170180220A1 (en) | Techniques to Generate Workload Performance Fingerprints for Cloud Infrastructure Elements | |
| CN104601696B (en) | Service processing method, service calling system, device and system | |
| CN104572301B (en) | A kind of resource allocation methods and system | |
| CN106330769A (en) | Business processing method and server | |
| CN103559093B (en) | The collocation method of a kind of server resource and device | |
| CN101727394B (en) | Method and device for testing performance of memory device | |
| CN102779075A (en) | Method, device and system for scheduling in multiprocessor nuclear system | |
| CN104683408A (en) | Method and system for OpenStack cloud computing management platform to build virtual machine instance | |
| CN107491346A (en) | A kind of task processing method of application, apparatus and system | |
| CN109831520A (en) | A kind of timed task dispatching method and relevant apparatus | |
| CN106886450A (en) | Method for scheduling task and system | |
| CN109710414A (en) | A kind of job scheduling method, device, equipment and storage medium | |
| CN107220102A (en) | The dormancy of virtual machine and awakening method and system | |
| CN103077135A (en) | Test resource management method capable of supporting parallel tests | |
| CN102855173A (en) | Method and device for testing software performance | |
| CN102457578A (en) | Distributed network monitoring method based on event mechanism | |
| CN104348852B (en) | A kind of method, apparatus and system for realizing telecommunication capability mass-sending | |
| CN106897133A (en) | A kind of implementation method based on the management cluster load of PBS job schedulings | |
| CN106385330A (en) | Network function virtualization composer realization method and device | |
| CN103927244A (en) | Plug-in scheduling process monitoring method implemented based on dynamic proxy | |
| CN105260297A (en) | Test data management system and method | |
| CN102546423B (en) | Method and device for queue scheduling and network device | |
| CN109800074A (en) | Task data concurrently executes method, apparatus and electronic equipment | |
| CN1652616B (en) | Adaptive system managing method | |
| CN110321204A (en) | Computing system, hardware accelerator management method and device and storage medium |
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 | ||
| CI01 | Correction of invention patent gazette |
Correction item: Claims Correct: 8 items False: 7 items Number: 15 Volume: 26 |
|
| CI03 | Correction of invention patent |
Correction item: Claims Correct: 8 items False: 7 items Number: 15 Page: Description Volume: 26 |
|
| ERR | Gazette correction |
Free format text: CORRECT: CLAIM OF RIGHT; FROM: ITEM 7 TO: ITEM 8 |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100414 Termination date: 20180207 |