CN100361081C - Method for processing multi-thread, multi-task and multi-processor - Google Patents
Method for processing multi-thread, multi-task and multi-processor Download PDFInfo
- Publication number
- CN100361081C CN100361081C CNB2005100018627A CN200510001862A CN100361081C CN 100361081 C CN100361081 C CN 100361081C CN B2005100018627 A CNB2005100018627 A CN B2005100018627A CN 200510001862 A CN200510001862 A CN 200510001862A CN 100361081 C CN100361081 C CN 100361081C
- Authority
- CN
- China
- Prior art keywords
- thread
- tasks
- processors
- task
- processor
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
- G06F11/0754—Error or fault detection not based on redundancy by exceeding limits
- G06F11/0757—Error or fault detection not based on redundancy by exceeding limits by exceeding a time limit, i.e. time-out, e.g. watchdogs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
- G06F11/0715—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment in a system implementing multitasking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
- G06F11/0721—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment within a central processing unit [CPU]
- G06F11/0724—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment within a central processing unit [CPU] in a multiprocessor or a multi-core unit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0793—Remedial or corrective actions
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The present invention discloses a method for processing multiple threads, multiple tasks and multiple processors. First of all, the present invention checks whether threads / tasks / processors are in a running state or not in use, and then, exception treatment is carried out for the threads / the tasks / the processors which are not in the running state. When the present invention is used for monitoring the threads / processes during running, a step is added in order to judge whether the monitored threads and the tasks are in the running state or not, and the current states of the threads / the processes can be known via the step in order to further work out corresponding treatment.
Description
Technical field
The present invention relates to computer realm, be specifically related in the computing machine list/multiprocessor for the disposal route of multithreading/multitask.
Background technology
Along with the growth at full speed of the operational speed of a computer, the computer system of a lot of multiprocessors, multithreading, multitask has appearred, promptly by a plurality of processor concurrent workings, perhaps a plurality of independently tasks or thread parallel work in a processor.Present computer operating system mostly is to support multitask, based on the multitask space that this function provided, and programmer's operation of each fragment in the controlling application program fully, thus write out application program efficiently.
Usually, for multicomputer system, have the processor of a plurality of parallel runnings, it generally is infinite loop that the task of the limit priority in each processor is handled, and for example constantly checks each interrupt source, timer or interface message etc.; For the situation of uniprocessor multithreading, its each thread/task generally also is an infinite loop.The method of handling at present multithreading/multitask is, the system clock of each processor of systems inspection judges whether to arrive working time of each interrupt source, timer or interface message, and operation simultaneously arrives a plurality of thread/tasks of working time then.
But when a plurality of threads/a plurality of tasks/when a plurality of processors moved simultaneously, often appearance influenced the situation that other threads/task is carried out owing to certain thread/task breaks down.Task or thread generally have dormant state, run mode, pending state and ready attitude etc., operating system or other software can judge that perhaps certain tasks/threads is in certain state, but except that running status, other state can't judge whether it can arrive normal operating condition.If because some reason, some thread or task again can't be moved, and do not have corresponding means to detect.Therefore the state of each thread/task can't be in time understood, thereby corresponding processing can't be made timely.
Summary of the invention
The technical problem to be solved in the present invention is: a kind of method of handling multithreading/multitask is provided, can in time monitors the duty of each thread/tasks/processors, so that the normal operation of the assurance system that takes appropriate measures.
For solving the problems of the technologies described above, the invention provides a kind of method of handling multithreading/multitask/multiprocessor, comprise step:
1) in thread/tasks/processors arrival time of running and when bringing into operation, stored system time in the storer into, highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage, judges whether the system time of current thread/tasks/processors storage is identical with its system time of last time storing; If identical, judge that then this thread/tasks/processors is not in running status;
2) the described thread/tasks/processors that will not be in running status is carried out abnormality processing.
In addition, described step also comprises: monitored thread/tasks/processors is upgraded described system time.
Wherein, step 2) described abnormality processing comprises: produce alarm, the processor that breaks down of resetting, kill and restart fault thread/task/multiprocessor.
The present invention also provides a kind of method of handling multithreading/multitask/multiprocessor, comprises step:
1) in thread/tasks/processors arrival time of running and when bringing into operation, stored number of run in the storer into, highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage, judges whether the number of run of current thread/tasks/processors storage is identical with its number of run of last time storing; If identical, judge that then this thread/tasks/processors is not in running status;
2) the described thread/tasks/processors that will not be in running status is carried out abnormality processing.
In addition, described step also comprises: monitored thread/tasks/processors is upgraded described number of run.
Step 2 wherein) described abnormality processing comprises: produce alarm, the processor that breaks down of resetting, kill and restart fault thread/task/multiprocessor.
Compared with prior art, the invention has the beneficial effects as follows: the present invention is when each run monitoring level thread/process, increased and judged whether monitored thread, task are in the step of running status, by this step, can in time recognize the current state of each thread/process, and then make corresponding processing.
In addition, the present invention is by preserving the system time of each run thread/task in thread with highest priority/task, perhaps the method for the number of run by preserving thread/task in thread with highest priority/task judges whether the thread/task of arriving working time is in running status.If utilize the method for in thread with highest priority/task, preserving the system time of each run thread/task to detect the duty of each thread/task, then when breaking down, can learn the system time when breaking down, if a plurality of modules break down, can know the time sequencing of fault whereby, thereby make things convenient for positioning problems.
Description of drawings
Fig. 1 is the process flow diagram of thread/tasks/processors monitored among the present invention;
Fig. 2 is the workflow synoptic diagram of monitoring level thread/tasks/processors among the present invention.
Embodiment
In the computing machine in modern times, the processing speed of CPU is more and more faster, but because physical restriction, the read or write speed of external unit even internal memory is well below the speed of CPU, under therefore a lot of situations, need CPU to wait for after certain is operated and just can finish subsequent operation, in this process, can move other thread or task, the switching of thread or task occur.If because some reason (hardware fault or software issue), thread or task can not switch back again, so, will have thread or task to be in the state of hanging.
When there is the program example of a plurality of independent operatings that switched by hardware in processor, be called the system of multithreading, each thread is shared a processor, but is to use different hardware resources, promptly moves different contexts; When having only a processor, and can only switch by operating system the time, be called the system of multitask, processor of each task sharing, but can only move an example simultaneously; Situation for there being independent CPUs on a plurality of hardware belongs to distributed multi-processor system, primary processor is arranged generally and from the branch of processor, each processor can be multitask or multithreading working method.Primary processor need be monitored the work of other processors.The invention provides a kind of method of handling multithreading/multitask/multiprocessor, can in time detect the duty of each processor or thread/task, so that the normal operation of the assurance system that takes appropriate measures.
Use when of the present invention, check at first whether thread/tasks/processors is in running status; Described thread/the tasks/processors that will not be in running status is then carried out abnormality processing.
Limit priority task or thread or processor are monitored other tasks/threads/processors among the present invention.Check whether thread/tasks/processors arrives the time of running.The mode that generally can adopt timer regularly to trigger, the method that whether is in running status for the thread/tasks/processors that judge to arrive the time of running can be limit priority tasks/threads/processor distribution one segment memory, (fixed in the enough big time according to actual conditions, need to consider the dispatching cycle of each thread or task, for example 1 second), check whether the memory value of each thread/tasks/processors correspondence changes; If judged result is for being, then each thread/tasks/processors operation normally; Otherwise, can judge the operation exception of thread/tasks/processors.For the reliability that guarantees to judge, can continuous detecting N time (for example 3), if N continuous time does not all change, can confirm thread/tasks/processors fault, should carry out abnormality processing.
The invention provides two kinds and judge whether monitored thread/task is in the method for running status, a kind ofly be: be monitored thread/tasks/processors allocate memory; Monitored thread/tasks/processors deposits described system time in above-mentioned described storer; Be that highest priority thread/tasks/processors is distributed temporary storage then; Highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage, then current time and the system time of preserving are compared, whether identical by judging the system time that current time and thread/tasks/processors last time preserved, judge whether this thread/task is in running status.Another kind method is: be monitored thread/tasks/processors allocate memory; Monitored thread/tasks/processors deposits described number of run in above-mentioned described storer; For highest priority thread/tasks/processors is distributed temporary storage; Highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage; According to current number of run and the number of run of preserving, judge whether this thread/task is in running status then.
When detecting fault, then carry out abnormality processing, abnormality processing comprises: produce alarm, the processor that breaks down of resetting, kill and restart fault thread/task.
Below be the concrete principle of work that the present invention relates to:
In general, for multicomputer system, have the processor of a plurality of parallel runnings, the task handling of the limit priority in each processor generally all is an infinite loop, for example constantly checks each interrupt source, timer or interface message or the like; For the situation of uniprocessor multithreading, its each thread/task generally also is an infinite loop.
In order to check the state of each processor or thread, the present invention has added the step whether thread/task of judging current time is in running status in circulation, will be described in detail in conjunction with first embodiment and second embodiment for concrete deterministic process.
First embodiment: in the present embodiment, system clock is write the specific memory device, judge according to the system clock of preserving in the storer whether system time changes.Concrete step is as described below.
At first distributing a segment memory in public internal memory, is each monitored thread/tasks/processors allocate memory then, is that the thread/tasks/processors of monitoring level is distributed temporary storage (step 1) simultaneously;
Fig. 1 is the particular flow sheet of each detected thread/tasks/processors in the present embodiment.In the present embodiment, the course of work of the thread/tasks/processors of each monitored level is: when each detected thread/tasks/processors arrives the time of running and when bringing into operation, each detected thread/tasks/processors is obtained current system time, and this time is written as in the storer of its distribution, every operation once, the system time of preserving in the storer is once upgraded, carry out normal process again.
Fig. 2 is the process flow diagram of the thread/tasks/processors of monitoring level for the thread/tasks/processors of limit priority in the present embodiment.When each detected thread/tasks/processors arrived the time of running, the thread/tasks/processors of limit priority read the system time of storing in the storer of each thread/tasks/processors correspondence, and writes temporary storage (step 2); Consideration is kept at system time in other processor of limit priority or the thread/task temporarily is used for follow-up comparison.Because the processing of all tasks/threads/processors generally all is an infinite loop, therefore, in case certain adjacent 2 circulation, the system time of reading remains constant, illustrates that then this processor or thread/task may break down.For the reliability that guarantees to judge, can continuous detecting N time, therefore can set certain delay time, for example 1000ms (step 3); Constantly read the updating value (step 4) that the storer of thread/tasks/processors correspondence obtains in period at this section; Judge whether updating value equals the preceding value of once reading (step 5) of temporary storage storage,, then upgrade the value (step 6) of temporary storage if different; If identical, then carry out abnormality processing (step 7), judge whether to have traveled through all thread/tasks/processors (step 8) then; If no, then continue to check whether next thread/task arrives the time of running, and judge (step 9) according to step 5 is described;
Behind all thread/tasks of traversal or processor, move normal thread/task (step 10) that all arrive the time of running.
Duty for thread/task comprises: sleep state, execution state, ready attitude, pending state, dead attitude etc.An advantage of present embodiment is owing to preserved the system time of each thread/task run, therefore not only can in time understand the running status of each processor or thread/task, and when breaking down in system, can understand the module that fault occurs, and the time sequencing that breaks down of each module and concrete fault-time, made things convenient for the location of problem.
Concrete position fixing process is as described below: the different characteristics according to different duties preestablishes certain time period, judges the present residing state of this thread/task according to this time period then.If processor or thread/task that primary processor or monitoring thread/task (being the highest processor of priority level or thread/task) discovery exists system time no longer to change then will be carried out abnormality processing.Concrete abnormality processing mode can produce the processor of the alarm or the fault that resets or kill the fault thread and restart this thread.
Second embodiment: in the present embodiment, in the specific memory device, counter is set, judges according to the number of run of writing down in the storer of thread/task whether thread/the task of arriving working time is in running status.Concrete step is as described below.
In the present embodiment, at first in public internal memory, distribute a segment memory, distribute different addresses separately for each processor or thread/tasks/processors then, and in each storer, counter is set;
When detected thread/tasks/processors arrives the time of running and when bringing into operation, in the counter of storer that is each detected processor or thread/Task Distribution, write down the number of run of this detected thread/tasks/processors; Other thread/tasks/processors of monitoring level reads the number of run of storing in each monitored thread/tasks/processors storer then, and is saved in the temporary storage.In like manner, it is because the processing of highest level generally all is an infinite loop that consideration is kept at number of run in other processor of limit priority or the thread/task, therefore, in case certain circulation or certain section interior circulation of time, number of run does not change, and illustrates that then this processor or thread/task break down.
Judge whether current number of run is identical with the preceding number of run of once preserving; Judge according to number of run whether current thread/task is in running status in the present embodiment, because the every operation of thread/process once, therefore the value of counter will change, if number of run remains constantly, the not operation in certain period of this thread/process will be described then;
If inequality, then further judge whether to have traveled through all thread/tasks or processor, when not having, continue to check whether next thread/task arrives the time of running, and judge according to above-mentioned steps;
Otherwise, if identical, then carry out abnormality processing, continue then to check whether next thread/task arrives the time of running, and judge according to above-mentioned steps;
Behind all thread/tasks of traversal or processor, move normal thread/task that all arrive the time of running.
Can carry out positioning problems equally according to the number of run that writes down in the storer in the present embodiment.Concrete position fixing process is as described below: the different characteristics according to different duties preestablishes certain time period, judges the present residing state of this thread/task according to this time period then.If primary processor or monitoring thread/task (being the highest processor of priority level or thread/task) are found to have the processor or the thread/task of not reruning, then will carry out abnormality processing.Concrete abnormality processing mode can produce the processor of the alarm or the fault that resets or kill the fault thread and restart this thread.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (6)
1, a kind of method of handling multithreading/multitask/multiprocessor is characterized in that, comprises step:
1) in thread/tasks/processors arrival time of running and when bringing into operation, stored system time in the storer into, highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage, judges whether the system time of current thread/tasks/processors storage is identical with its system time of last time storing; If identical, judge that then this thread/tasks/processors is not in running status;
2) the described thread/tasks/processors that will not be in running status is carried out abnormality processing.
2, the method for processing multithreading/multitask according to claim 1/multiprocessor is characterized in that, described step also comprises: monitored thread/tasks/processors is upgraded described system time.
3, the method for processing multithreading/multitask according to claim 1 and 2/multiprocessor is characterized in that step 2) described abnormality processing comprises: produces alarm, the processor that breaks down of resetting, kills and restart fault thread/task/multiprocessor.
4, a kind of method of handling multithreading/multitask/multiprocessor is characterized in that, comprises step:
1) in thread/tasks/processors arrival time of running and when bringing into operation, stored number of run in the storer into, highest priority thread/tasks/processors reads the storer of monitored thread/tasks/processors correspondence to temporary storage, judges whether the number of run of current thread/tasks/processors storage is identical with its number of run of last time storing; If identical, judge that then this thread/tasks/processors is not in running status;
2) the described thread/tasks/processors that will not be in running status is carried out abnormality processing.
5, the method for processing multithreading/multitask according to claim 4/multiprocessor is characterized in that, described step also comprises: monitored thread/tasks/processors is upgraded described number of run.
6, according to the method for claim 4 or 5 described processing multithreading/multitask/multiprocessors, it is characterized in that step 2) described abnormality processing comprises: produces alarm, the processor that breaks down of resetting, kills and restart fault thread/task/multiprocessor.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100018627A CN100361081C (en) | 2005-01-18 | 2005-01-18 | Method for processing multi-thread, multi-task and multi-processor |
| PCT/CN2006/000021 WO2006076853A1 (en) | 2005-01-18 | 2006-01-09 | A method of processing multi-thread/multi-task/multi-processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100018627A CN100361081C (en) | 2005-01-18 | 2005-01-18 | Method for processing multi-thread, multi-task and multi-processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1808386A CN1808386A (en) | 2006-07-26 |
| CN100361081C true CN100361081C (en) | 2008-01-09 |
Family
ID=36691981
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100018627A Expired - Fee Related CN100361081C (en) | 2005-01-18 | 2005-01-18 | Method for processing multi-thread, multi-task and multi-processor |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN100361081C (en) |
| WO (1) | WO2006076853A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105094974A (en) * | 2015-08-14 | 2015-11-25 | 上海斐讯数据通信技术有限公司 | Burst-mode signal processing method and system |
| CN109933459A (en) * | 2017-12-18 | 2019-06-25 | 北京京东尚科信息技术有限公司 | A kind of execution method and apparatus of multitask |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622291A (en) * | 2012-03-13 | 2012-08-01 | 苏州阔地网络科技有限公司 | Method and system for monitoring processes |
| CN107713992B (en) * | 2013-03-15 | 2021-10-26 | 雅培糖尿病护理公司 | Data processing system |
| CN103425545A (en) * | 2013-08-20 | 2013-12-04 | 浪潮电子信息产业股份有限公司 | System fault tolerance method for multiprocessor server |
| WO2017070900A1 (en) * | 2015-10-29 | 2017-05-04 | 华为技术有限公司 | Method and apparatus for processing task in a multi-core digital signal processing system |
| CN105912416B (en) * | 2016-04-07 | 2019-06-28 | 珠海市魅族科技有限公司 | A kind of method and terminal monitoring processor in the terminal |
| CN108243056B (en) * | 2016-12-27 | 2020-11-20 | 大唐移动通信设备有限公司 | Method and device for acquiring abnormal information |
| CN109271268B (en) * | 2018-09-04 | 2022-03-18 | 超越科技股份有限公司 | DPDK-based intelligent fault tolerance method |
| CN110335150A (en) * | 2019-06-26 | 2019-10-15 | 苏州工业园区服务外包职业学院 | A kind of host processing system of transaction system platform service terminal |
| CN111143127B (en) * | 2019-12-23 | 2023-09-26 | 杭州迪普科技股份有限公司 | Method, device, storage medium and equipment for supervising network equipment |
| CN112469065A (en) * | 2020-12-23 | 2021-03-09 | 北京春笛网络信息技术服务有限公司 | Keep-alive detection method for 5G short message all-in-one machine |
| CN114115156B (en) * | 2021-11-26 | 2024-01-30 | 重庆理工大学 | Rescheduling method for flexible job shop |
| CN116521351B (en) * | 2023-07-03 | 2023-09-05 | 建信金融科技有限责任公司 | Multithreading task scheduling method and device, storage medium and processor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1319203A (en) * | 1998-07-22 | 2001-10-24 | 艾利森电话股份有限公司 | Method relating to handling of transactions in databases |
| CN1469243A (en) * | 2003-06-24 | 2004-01-21 | 北京邮电大学 | Task assigning mechanism for large-scale distributive invasion detecting system |
| CN1489049A (en) * | 2002-10-11 | 2004-04-14 | 华为技术有限公司 | Method for monitoring background programme |
| CN1539105A (en) * | 2001-08-03 | 2004-10-20 | ���Ĺ��ʹ�˾ | Victim selection for deadlock detction |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001075802A (en) * | 1999-09-08 | 2001-03-23 | Fujitsu Ltd | Speculative execution device and verification device |
| FR2816730B1 (en) * | 2000-11-13 | 2004-10-15 | Commissariat Energie Atomique | SECURITY METHOD FOR DETERMINING REAL-TIME EXECUTION OF MULTITASTIC APPLICATIONS OF CONTROL-CONTROL TYPE WITH ERROR CONTAINMENT |
-
2005
- 2005-01-18 CN CNB2005100018627A patent/CN100361081C/en not_active Expired - Fee Related
-
2006
- 2006-01-09 WO PCT/CN2006/000021 patent/WO2006076853A1/en not_active Application Discontinuation
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1319203A (en) * | 1998-07-22 | 2001-10-24 | 艾利森电话股份有限公司 | Method relating to handling of transactions in databases |
| CN1539105A (en) * | 2001-08-03 | 2004-10-20 | ���Ĺ��ʹ�˾ | Victim selection for deadlock detction |
| CN1489049A (en) * | 2002-10-11 | 2004-04-14 | 华为技术有限公司 | Method for monitoring background programme |
| CN1469243A (en) * | 2003-06-24 | 2004-01-21 | 北京邮电大学 | Task assigning mechanism for large-scale distributive invasion detecting system |
Non-Patent Citations (2)
| Title |
|---|
| 一种简单的线程死锁检测方法及其VC++应用. 郑冬黎.湖北汽车工业学院学报，第18卷第2期. 2004 * |
| 检测并杀死失去响应的进程. 陈锐.电脑知识与技术. 2003 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105094974A (en) * | 2015-08-14 | 2015-11-25 | 上海斐讯数据通信技术有限公司 | Burst-mode signal processing method and system |
| CN109933459A (en) * | 2017-12-18 | 2019-06-25 | 北京京东尚科信息技术有限公司 | A kind of execution method and apparatus of multitask |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1808386A (en) | 2006-07-26 |
| WO2006076853A1 (en) | 2006-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100361081C (en) | Method for processing multi-thread, multi-task and multi-processor | |
| US7426657B2 (en) | System and method for predictive processor failure recovery | |
| US7631307B2 (en) | User-programmable low-overhead multithreading | |
| US9092023B2 (en) | Industrial controller using shared memory multicore architecture | |
| US7992042B2 (en) | Debug support device, and program for directing computer to perform debugging method | |
| US7120784B2 (en) | Thread-specific branch prediction by logically splitting branch history tables and predicted target address cache in a simultaneous multithreading processing environment | |
| US8190864B1 (en) | APIC implementation for a highly-threaded x86 processor | |
| CN100338555C (en) | Method and apparatus for reducing power consumption in a logically partitioned data processing system | |
| EP1839146B1 (en) | Mechanism to schedule threads on os-sequestered without operating system intervention | |
| US6996745B1 (en) | Process for shutting down a CPU in a SMP configuration | |
| US20080148259A1 (en) | Structured exception handling for application-managed thread units | |
| US9354934B2 (en) | Partitioned shared processor interrupt-intensive task segregator | |
| US20170147422A1 (en) | External software fault detection system for distributed multi-cpu architecture | |
| US11645081B2 (en) | Handling exceptions in a multi-tile processing arrangement | |
| JPWO2009150815A1 (en) | Multiprocessor system | |
| JP2019114173A (en) | Information processing apparatus, information processing method, and program | |
| JP2009048358A (en) | Information processor and scheduling method | |
| JPS59167756A (en) | Dispatch control system of virtual computer | |
| JP4759026B2 (en) | Processor | |
| KR20080089564A (en) | Centralized interrupt controller | |
| JP5299681B2 (en) | Program inspection method | |
| US20240134669A1 (en) | Paravirtual pause loops in guest user space | |
| JP2008077388A (en) | Multiprocessor control system, method and program | |
| JP2004318658A (en) | Debugging control device and control method thereof | |
| CN115858107A (en) | Multithreading batch operation monitoring method and device, computer equipment 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 | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160427 Address after: American California Patentee after: Snaptrack, Inc. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: Huawei Technologies Co., Ltd. |
|
| 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: 20080109 Termination date: 20210118 |