WO2006092318A1 - Verfahren zur echtzeitanalyse eines systems - Google Patents
Verfahren zur echtzeitanalyse eines systems Download PDFInfo
- Publication number
- WO2006092318A1 WO2006092318A1 PCT/EP2006/001955 EP2006001955W WO2006092318A1 WO 2006092318 A1 WO2006092318 A1 WO 2006092318A1 EP 2006001955 W EP2006001955 W EP 2006001955W WO 2006092318 A1 WO2006092318 A1 WO 2006092318A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- act
- time
- task
- time interval
- approxlist
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
- G06F9/4887—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues involving deadlines, e.g. rate based, periodic
Definitions
- the invention relates to a method for real-time analysis of a system, in particular a technical system.
- tasks In a variety of technical systems, such as embedded real-time systems or embedded computer systems, it is necessary that the system processes tasks, so-called tasks, in a predetermined time interval, which is also designated with a time limit or deadline. H. is real-time capable. Tasks to be performed by the system are also summarized below under a "task system".
- Real-time analysis can be used to automate the development of real-time systems. Furthermore, it is possible to perform the real-time analysis while the task system is being executed by the system. If it is determined during execution that specified real-time conditions can not be met by the system, then z. B. measures for limited but still largely secure processing of the task system can be initiated. Such measures can z. For example, outsourced individual tasks for their processing on other systems or system components. It is also possible to defer the execution of individual tasks in favor of other tasks.
- Real-time analysis methods can be classified into adequate and necessary procedures. With a sufficient procedure, a system can always be classified error-free for a given task system. On the other hand, it may be possible with a sufficient procedure that the system is incorrectly classified as non-real-time capable for a given task system, although the system actually would be able to process the task system in compliance with the specified real-time conditions.
- a necessary method for real-time analysis for a processor of a computer is known.
- a maximum test interval i. H. a maximum time interval calculated for a given task system.
- a time interval located in the test interval is selected according to a predetermined criterion, and the computing time required by the processor for processing the tasks to be processed in the time interval is calculated. Then the calculation time is compared with the length of the time interval.
- a task system can not be processed in real time if the computing time is greater than the length of the time interval. If the system is real-time capable for the selected time interval, another time interval located in the test interval is checked. The process continues until either it is determined that the system is not real-time capable or until all of the time intervals satisfying the criterion have been verified. Although in the method only time intervals need to be checked, which are within the test interval. However, it may be that many time intervals are unnecessarily checked, unnecessarily prolonging the running time of the process.
- the computation time is compared with a system capacity, which is available to the system in the next smaller time interval for the execution of tasks. Based on the comparison, it is determined whether the system is real-time capable.
- a disadvantage of the method is that the comparison leads to an error which can not be completely compensated even by a high number of time intervals. Many systems are falsely classified as non-real-time capable for a given task system.
- a further disadvantage is that, given a small maximum error selected, the running time of the method can be greater than that of a precise method. In the above methods known in the art, it is necessary to set an accuracy or error of the method before performing the method. An error that is too large for a given task system can cause the system to be incorrectly classified as non-real-time capable. A too small selected error leads to a high expenditure and to high running times.
- the object of the invention is to eliminate the disadvantages of the prior art.
- a particularly fast and accurate method for real-time analysis should be specified.
- This object is achieved by the features of claims 1 and 26 to 28.
- Advantageous embodiments of the invention will become apparent from the features of claims 2 to 25.
- a job cost defined by the execution of a task causes the following steps:
- the time interval is considered to be processable in real time if the total system costs do not exceed the limit, wherein (ii) if the total system cost exceeds the limit, the second quantity, if not empty, is reduced such that system cost is taken into account for further determination of the total system cost for at least one task of a second set of approximate value task and at least that Steps lit. c) and d) are carried out repeatedly, and wherein
- an approximate value is used for a job of the first quantity and an approximate value for a job of the second quantity.
- the approximate system cost of a job can be determined based on the system cost of the job.
- the system costs can be a processing time or computing time or a utilization of the system.
- the method is an approximation method.
- the accuracy of the method depends on the thickness of the second set.
- the accuracy or "degree of approximation" of the method is a measure of the deviation of the approximation method from an exact method. As the degree of approximation increases, the effort to perform the approximation method, e.g. B. the term.
- the degree of approximation can be changed dynamically during the method.
- An increase in the degree of approximation can be achieved by reducing the second amount in step lit. d) ii).
- a reduction / enlargement of the first or second quantity is understood to mean a reduction / increase in the thickness of the first or second quantity.
- the thickness of the second quantity can already be selected as large as possible at the beginning of the process.
- the degree of approximation can be dynamically increased during the process by reducing the second amount.
- a reduction in the second quantity means that the system costs are used to determine the total system costs for at least one job of the second quantity instead of the approximate value. Ie. at least one job of the second quantity is assigned to the first quantity, which is thereby increased. If the steps lit. c) and d) the system cost is used instead of the approximate value of the job. Because the approximate value is greater than the system cost, the overall system cost can be reduced by an approximate error. Due to the reduction of the total system cost, it may be possible that these are smaller than the limit. In this case, the task system may be classified as being processable by the system in real time for the time interval being checked. If the total system costs are greater than the limit despite the reduction, the second quantity can be further reduced and the first quantity can be further increased. Subsequently, the
- the system is in particular a technical system, for.
- a technical system for.
- an embedded real-time or computer system with one or more interconnected computers, electronic circuits and controls, and the like.
- the second amount when the total system cost is less than the threshold, the second amount is increased. Increasing the second set increases the number of approximated jobs. Consequently, the degree of approximation decreases and the running time of the method can be reduced. The process can be carried out even faster and more efficiently.
- At least one task is assigned a changeable test limit, whereby for tasks whose test limit is exceeded in the time interval, the approximate value is used for the jobs of the task, and wherein if the total system costs are greater than the limit, at least one test limit of a task is changed such that the second amount is reduced.
- the ratio of the approximate values to the total system costs can be limited or shifted.
- the ratio can z. B. used to determine a current error or a current degree of approximation and passed to the user.
- the thickness of the second set and the associated degree of approximation can be changed in a particularly simple manner by increasing or decreasing the test limits. By increasing the test limit / n, the degree of approximation can be increased.
- the test limit for ne task is a number of jobs of the task.
- the test limit is reached if there are a corresponding number of jobs in the time interval of the number. It is particularly advantageous, at least initially, to choose the same test limit for all tasks, eg. For example, the same number of jobs.
- At least one numerical value such.
- B. a test limit, increased, preferably doubled, and / or to increase the second amount at least one numerical value, such.
- a change in the numerical value can be done, for example, by a predefined arithmetic operation in a particularly simple and fast manner.
- the method is continued with an already checked time interval in which the overall system costs were less than the limit value.
- an arbitrary time interval can be selected in which the total system costs were less than the limit value.
- results of already checked time intervals can be used. A repeated implementation of the method from the beginning can be avoided. The process can be carried out particularly quickly and efficiently.
- one of the following characteristics of the tasks is used to reduce the second quantity takes into account: amount of actual system cost, error rate of approximate system cost of a total error of the procedure. It is possible to first eliminate the approximation for jobs that cause the highest system cost. Similarly, the approximation can first be reversed for those jobs that have the largest error fraction. This can be achieved that the total system costs are reduced by the highest possible amount.
- the real-time capability is checked successively for time intervals of increasing size. If a time interval can be processed in real time, all smaller time intervals can also be considered to be processable in real time. If the total system costs exceed the limit value in a checked time interval, the method can be continued in the same time interval with a higher degree of approximation. If the time intervals are checked in ascending order of magnitude, the total system costs can be particularly simple, eg. B. by adding up the system costs and the approximate values are determined.
- the tasks for the determination of the total system costs are grouped. Such a procedure is particularly useful if they are at different priority levels.
- At least one task is repeatedly executed by the system.
- at least one task is executed with a minimum time interval or periodically with one period.
- a further embodiment of the invention provides that the system costs are calculated on the basis of a processing time required for the execution of the tasks. It it is also possible that the system costs are calculated on the basis of an upper limit for the maximum required processing time. Furthermore, it is possible to calculate the system cost on the basis of a utilization of an execution component, preferably a CPU, of the system required for the execution of the jobs.
- the limit value is determined on the basis of a capacity of the system available in the time interval. For example, it is possible to
- Limit using the capacity of a processor or a CPU to determine If a constant capacity is assumed, the limit can be determined by multiplying the length of the time interval by the capacity. In this determination, a time interval of twice the length is assigned the double limit. It is also possible to describe the limit by an amount of system cost that can be processed by the CPU within the time interval associated with the limit. Furthermore, different limit values can be used for different time intervals. This allows better modeling of processors with fluctuating capacity. It is also possible to assign different limit values only to some time intervals. The limit values are preferably assigned in increasing size to time intervals of likewise increasing size. A threshold determined for a time interval may be used to determine the threshold of a larger, e.g. B. the next larger, time interval can be used. Compared with the method with fixed limits, a much more precise analysis can be made possible with a small additional calculation effort.
- the total system costs for discrete, an initial and an end time having time intervals are determined, the end time
- An end of a time limit is up to which a job of a task is to be processed at the latest.
- the approximate value for a job is calculated based on a specific utilization of the system.
- the specific utilization in turn can be calculated as a quotient of the processing time and period of the task. It is also possible that the specific utilization is taken into account for an interval which is smaller than the time interval.
- That task is processed by the system first, for which the end of the time limit is closest in time. It is also possible for the tasks to be processed by the system in a priority order. Furthermore, the tasks can be described with an event stream model.
- R act R act + C act + ( act - I old ) * U approx WHILE (R act > I act )
- ⁇ rev means the number of tasks for which the system costs are taken into account in a further run of the algorithm instead of the respective approximate values.
- a computer program product comprising computer readable program code means for performing the method according to the invention on a
- Computer system provided. Furthermore, a digital storage medium with a computer-readable program stored thereon for carrying out the method according to the invention is provided. Furthermore, a computer system is provided, comprising a digital storage medium with program code means executable by the computer system for carrying out the method according to the invention.
- an approximation method for real-time analysis of a system is considered.
- a system-executable task system ⁇ n with n tasks ⁇ is provided, where n is a natural number.
- system costs which caused by the execution of a task ⁇ € ⁇ n
- a computation time required by the system for executing the task ⁇ is used.
- Time intervals I are checked which are smaller than a predefined maximum time interval I max .
- the time intervals I have a common start time.
- An end time of a time interval I is given by an end of a time limit d ⁇ of a job of a task ⁇ of the task system ⁇ n . Ie. the job must be completed at the latest by the end time.
- time intervals I ⁇ i he, ..., he I ⁇ n ⁇ e I max are selected.
- the time intervals I ⁇ ⁇ , ... or I ⁇ n include 1, 2, ... and n ⁇ jobs of the task ⁇ .
- a test limit T max ( ⁇ ) dependent on the parameters of the task ⁇ is also determined, for which applies: T max ( ⁇ ) he I max .
- T max ( ⁇ ) he I max When determining a total computing time, the actual computing time of the jobs of the task ⁇ is used for the time intervals I er T max ( ⁇ ).
- time intervals I For time intervals I, with T max ( ⁇ ) he I er I max an approximate value for the computing time is used, which is greater than the computing time.
- corresponding time intervals are selected for each task of the task system.
- the numbers n ⁇ and the error of the approximation method are interdependent.
- the selected time intervals are checked in ascending order.
- the total computing time is determined which is required by the system for processing the jobs arising therein and compared with a limit value for the computing time in the respective time interval.
- the system is real-time capable if the total computation time is less than the limit.
- n ⁇ 1 If the total computation time containing an error exceeds a limit for the computation time available in the checked time interval, the test limit becomes
- T max U is shifted or increased and, accordingly, the maximum number n t of the time intervals is increased for at least one task ⁇ .
- the real-time capability for the time interval is checked again by recalculating the total system costs and comparing them with the limit value. If the total system costs are less than the limit value, the next larger time interval can be checked. If the total system costs continue to exceed the limit value, once again one or more test limits T max ( ⁇ ) can be shifted.
- the test limit T max ( ⁇ ) can be z. B. be increased by doubling the number n ⁇ .
- the test limit T max ( ⁇ ) can be z. B. be increased by doubling the number n ⁇ .
- at least one test limit T max ( ⁇ ) is shifted over the end time of the time interval being checked. This affects the number of jobs for which an approximate calculation time value is used in the time interval. If the number of approximated jobs is reduced, the total computation time is reduced by the error rate of the jobs that are no longer approaching. As a result, by increasing the degree of approximation, the determined total computing time can be reduced. It may be that this reduction reduces the determined total computation time below the limit value. If this can be achieved, the task system can be accepted for the checked time interval.
- time intervals I are checked in size in ascending order. If the checked time interval I is classified as realizable in real time, this also applies to all smaller time intervals I. If the checked time interval I is classified as not processable in real time, the system costs are used instead of the approximate value for at least one approximated job of a task , It is sufficient to recheck the already checked time interval I from. It is not mandatory that the real-time analysis be continued with a smaller time interval I or repeated from the beginning.
- the method can be realized as follows:
- a specific utilization U of the task system ⁇ n can be calculated as follows:
- ⁇ ⁇ , i 1, 2, ... tasks from the task system ⁇ n , c ⁇ the system costs, such. For example, a computation time required by the system to execute the task ⁇ i, and pi a period or a minimum time interval of the task Xj . ,
- a first step it is checked whether the specific utilization U exceeds the value 1. In this case, the task system is not real-time capable and the real-time analysis can be ended.
- one or more maximum test intervals I max are determined. Furthermore, variables required to perform the method are initialized, such as. For example, the total system cost as a cumulative current calculation time R act for the current time interval I act - A time interval preceding the current time interval I act is denoted by I o i d .
- the respective first time interval I ⁇ i i of the task Ti of the task system T n is inserted into a test list TestList.
- the first time interval I ⁇ i i results from a time limit d ⁇ .
- an allowable number of time intervals is determined.
- the number Anz is set to 1 and a list ApproxList of tasks ⁇ , whose jobs are approached, is empty.
- the subsequent instructions are repeatedly executed until either the test list TestList is empty, or the maximum test interval is exceeded or reached, ie when, for example, the test list is empty. For example, I ac t ⁇ Imax
- the first instruction is the current time interval I act .
- the current time interval I act is equated with the smallest time interval from the test list TestList. The smallest time interval is removed from the test list TestList. If this time interval is contained several times for different tasks Ti in the test list TestList, only one entry is deleted. ⁇ ac t denotes the task associated with the current time interval I act .
- the calculation time R act is calculated using the following formula:
- ⁇ acfc R-old + C act + U appro ⁇ X (I aat - I 0 Id)
- U appro ⁇ denotes a specific utilization of the approached tasks.
- U approx can be calculated with the following formula:
- U apPro ⁇ can also be stored in a variable and updated during real-time analysis if necessary.
- the cumulative current computing time R act is compared with a limit value GW act for the current time interval I ac t.
- the limit value GW act can be equated, for example, with the interval length of the current time interval I act .
- the real-time detection for the time interval I act is provided.
- a next instruction checks whether the test limit T max ( ⁇ ac t) of the current task ⁇ ac t has been reached or exceeded.
- the test limit T max ( ⁇ act ) depends on the currently permissible number of times the time interval. Ie off. For periodic tasks, the check can be done by the following formula:
- d act denotes the current time limit and p act the period of the task ⁇ ac t. If the test limit T max ( ⁇ act ) is reached, the task ⁇ act is added to the list ApproxList of the approximated tasks in a next instruction. If necessary, U approx is updated.
- a next larger time interval I next for the task ⁇ act is inserted into the test list TestList.
- the time interval I next can be determined as follows:
- the value of the last accepted time interval I o i d is set to the value of the current time interval I ac t.
- the number is increased.
- the number Anz can be doubled, for example. It is then checked whether, due to the increase, one or more tasks included in the list ApproxList are no longer approached. The no longer approached tasks are determined. Those tasks are determined which, as a result of the increased number Anz, are assigned a test limit T max ( ⁇ ) which is greater than the current time interval I act . The tasks thus determined are removed from the list ApproxList and their next time interval, which is greater than I act , is inserted in the list TestList. Then the computation time R act is reduced by the error proportion of the tasks no longer approximated and as determined above. For periodic tasks, the error in a time interval I can be calculated as follows:
- ⁇ ⁇ r pi, and c ⁇ denote the time limit, period and system cost, eg.
- a task Ti of Tasksy- stems ⁇ n As the computing time, a task Ti of Tasksy- stems ⁇ n .
- the errors calculated in this way are subtracted from the calculation time R act . If the reduced computing time R ' act is still greater than the limit value GW act , then the increase in the number of pulses was not yet sufficient. The number Anz is further increased, for example doubled again. The increase in the number is repeated until either the reduced calculation time R ' act is less than or equal to the limit value GW a c t , or until the list ApproxList is empty.
- calculation time R ' act drops below the limit value GW act as a result of an increase in the number Anz, the method is continued as in the first case. If the calculation time R ' ac t is greater than the limit value GW act for every increase in the number Anz, then the system is not capable of real-time execution for the task system ⁇ n .
- the degree of approximation can be reduced again.
- the degree of approximation can be dynamically adapted to the requirements of the task system ⁇ n be adjusted, whereby the duration of the process can be further minimized. For example, a maximum number of tasks can be approached for each time interval. The following procedure can be used:
- the time intervals are checked in size in ascending order. Is determined is that the calculation time R act in a calibrated time interval is less than the limit value GW act, the number Num is reduced so that the number of tasks for which an approximate value for the
- test limit T max (O for at least one task ⁇ is shifted in such a way that the task ⁇ is approximated in the time interval I ac t or I new .
- the tasks ⁇ become as early as possible, e.g. B. each after the first time interval I ⁇ i, approximated. This approximation can be canceled as described above. If an approximation is removed, it may be attempted to approximate as soon as possible, z. B. in the next time interval to reintroduce. In the method, it is possible to insert time intervals only in the test list TestList, if this is due to the cancellation of an approximation. The method allows a far-reaching approximation, is fast and accurate.
- the fourth embodiment can be realized, for example, as follows:
- the first time intervals I ⁇ ⁇ which are produced in each case from the first job of each task ⁇ become the test list
- TestList inserted.
- the time intervals I ⁇ i in the test list TestList are then processed in ascending order of size, ie the real-time capability of the Systems checked for these intervals. All other time intervals I of the tasks are initially approximated, ie are not included in the test list TestList. If real-time detection fails for a time interval 1 ⁇ , the approximation is gradually canceled as described above. This is done either until the real-time evidence for the checked time interval I is provided or no task can be approached. If a task can no longer be approached, the task system is not real-time capable.
- the cancellation of the approximation of a task ⁇ leads to the fact that for this task ⁇ a time interval I ⁇ i, i> 2, is inserted into the test list TestList.
- the time interval I ⁇ i corresponds to the next larger time interval of the task ⁇ , which follows a time interval in which the test failed.
- the next larger time interval can be determined in the same way as the dynamic error real time analysis.
- the real-time analysis ends when all tasks are approached in a time interval and / or the calculation time R act is greater than the limit value GW ac t-
- a particularly low duration of the method can be achieved due to a significant reduction in the number of time intervals to be checked.
- z. B. dynamic it is possible to analyze the real-time capability of systems particularly accurate.
- the inventive method can by a, in particular technical, system such.
- a computer system are executed in parallel to the execution of the task system. It can be checked in advance whether the tasks of the task system can be processed in real time. If z. B. determined that a real-time condition for a given future date can not be met, it is possible individual tasks to outsource for faster processing to other system components or other systems. By such outsourcing can be achieved in the favorable case that the real-time condition can be met in the future time. It is also possible that a processing of individual tasks is postponed. For example, the processing of such tasks can be postponed, which are not mandatory for safe operation of the system or emergency operation of the system. It can improve the functioning of the system.
- the method according to the invention can also be used in a system with several interacting components.
- Information about the real-time capability of the components obtained by the method can be used to suitably plan or define a processing of tasks on the individual components. It can be optimized and improved the interaction of the components.
- inventive method for the development of technical systems such. Embedded real-time or computer systems, electronic circuits or controllers and the like. , be used.
- Ci costs d ⁇ , ie time limit
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2007557434A JP2008532150A (ja) | 2005-03-04 | 2006-03-03 | システムのリアルタイム分析のための方法 |
US11/884,916 US8185900B2 (en) | 2005-03-04 | 2006-03-03 | Method for the real-time capability analysis of a system by selectively using approximated or actual system expenses for jobs |
EP06723199A EP1854003A1 (de) | 2005-03-04 | 2006-03-03 | Verfahren zur echtzeitanalyse eines systems |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102005010580A DE102005010580A1 (de) | 2005-03-04 | 2005-03-04 | Verfahren zur Echtzeitanalyse eines Systems |
DE102005010580.7 | 2005-03-04 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006092318A1 true WO2006092318A1 (de) | 2006-09-08 |
Family
ID=36480896
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2006/001955 WO2006092318A1 (de) | 2005-03-04 | 2006-03-03 | Verfahren zur echtzeitanalyse eines systems |
Country Status (5)
Country | Link |
---|---|
US (1) | US8185900B2 (de) |
EP (1) | EP1854003A1 (de) |
JP (1) | JP2008532150A (de) |
DE (1) | DE102005010580A1 (de) |
WO (1) | WO2006092318A1 (de) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2396725A1 (de) * | 2009-02-16 | 2011-12-21 | Inchron GmbH | Verfahren zur analyse der echtzeitkapazität eines systems |
US9043644B2 (en) * | 2012-12-04 | 2015-05-26 | International Business Machines Corporation | Using separate processes to handle short-lived and long-lived jobs to reduce failure of processes |
US20150081400A1 (en) * | 2013-09-19 | 2015-03-19 | Infosys Limited | Watching ARM |
KR102264205B1 (ko) * | 2019-09-20 | 2021-06-10 | 인천대학교 산학협력단 | 실시간 시스템에서의 작업 할당 스케줄링이 가능한지 여부를 판정할 수 있는 실시간성 분석 장치 및 그 동작 방법 |
KR102252079B1 (ko) * | 2019-09-20 | 2021-05-13 | 인천대학교 산학협력단 | 실시간 시스템을 위한 실시간성 분석 장치 및 그 동작 방법 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2723653B1 (fr) * | 1994-08-11 | 1996-09-13 | Cegelec | Procede pour ordonnancer des taches successives qui ne subissent que des contraintes du type delais |
US5964829A (en) * | 1996-03-27 | 1999-10-12 | Lucent Technologies Inc. | Method and apparatus for providing enhanced pay per view in a video server employing a coarse-grained striping scheme |
GB9710522D0 (en) * | 1997-05-23 | 1997-07-16 | Rolls Royce Plc | Control system |
US7165252B1 (en) * | 1999-06-21 | 2007-01-16 | Jia Xu | Method of scheduling executions of processes with various types of timing properties and constraints |
JP2002342097A (ja) * | 2001-05-17 | 2002-11-29 | Matsushita Electric Ind Co Ltd | タスク割当可能時間決定装置及びタスク割当可能時間決定方法 |
-
2005
- 2005-03-04 DE DE102005010580A patent/DE102005010580A1/de not_active Ceased
-
2006
- 2006-03-03 WO PCT/EP2006/001955 patent/WO2006092318A1/de active Application Filing
- 2006-03-03 US US11/884,916 patent/US8185900B2/en active Active
- 2006-03-03 JP JP2007557434A patent/JP2008532150A/ja active Pending
- 2006-03-03 EP EP06723199A patent/EP1854003A1/de not_active Withdrawn
Non-Patent Citations (5)
Title |
---|
ALBERS K ET AL: "An event stream driven approximation for the analysis of real-time systems", REAL-TIME SYSTEMS, 2004. PROCEEDINGS. 16TH EUROMICRO CONFERENCE ON JUNE 30 - JULY 2, 2004, CATANIA, ITALY,IEEE, 30 June 2004 (2004-06-30), pages 187 - 195, XP002383592, ISSN: 1068-3070 * |
ALBERS K ET AL: "Efficient Feasibility Analysis for Real-Time Systems with EDF Scheduling", DESIGN, AUTOMATION AND TEST IN EUROPE, 2005. PROCEEDINGS MUNICH, GERMANY 07-11 MARCH 2005, PISCATAWAY, NJ, USA,IEEE, 7 March 2005 (2005-03-07), pages 492 - 497, XP010780009, ISBN: 0-7695-2288-2 * |
BARUAH S K ET AL: "Preemptively scheduling hard-real-time sporadic tasks on one processor", PROCEEDINGS OF THE REAL TIME SYSTEMS SYMPOSIUM. LAKE BUENA VISTA, DEC. 5 - 7, 1990, WASHINGTON, IEEE. COMP. SOC. PRESS, US, vol. SYMP. 11, 5 December 1990 (1990-12-05), pages 182 - 190, XP010022049, ISBN: 0-8186-2112-5 * |
CHAKRABORTY S., ET AL.: "Approximate Schedulability Analysis", PROCEEDINGS REAL-TIME SYSTEMS SYMPOSIUM ON 3-5 DECEMBER 2002, AUSTIN, TX, USA, IEEE, 3 December 2002 (2002-12-03), pages 1 - 10, XP002383591 * |
DEVI U C: "An improved schedulability test for uniprocessor periodic task systems", REAL-TIME SYSTEMS, 2003. PROCEEDINGS. 15TH EUROMICRO CONFERENCE ON 2-4 JULY 2003, PISCATAWAY, NJ, USA,IEEE, 2 July 2003 (2003-07-02), pages 23 - 30, XP010644814, ISBN: 0-7695-1936-9 * |
Also Published As
Publication number | Publication date |
---|---|
EP1854003A1 (de) | 2007-11-14 |
JP2008532150A (ja) | 2008-08-14 |
US20080276247A1 (en) | 2008-11-06 |
US8185900B2 (en) | 2012-05-22 |
DE102005010580A1 (de) | 2006-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE60226176T2 (de) | Verfahren und programme zur einstellung von prioritätsstufen in einem datenverarbeitungssystem mit multiprogrammierung und priorisierte warteschlangenbildung | |
EP1831786B1 (de) | Verfahren zur verteilung von rechenzeit in einem rechnersystem | |
EP0771444B1 (de) | Verfahren zur steuerung von technischen vorgängen oder prozessen | |
EP1756714A2 (de) | Verfahren zur prüfung der echtzeitfähigkeit eines systems | |
EP0635784A1 (de) | Multiprozessorsystem | |
EP1854003A1 (de) | Verfahren zur echtzeitanalyse eines systems | |
DE102007051803A1 (de) | Verfahren und Vorrichtung zur Datenverarbeitung | |
WO2006045734A1 (de) | Verfahren, betriebssystem und rechengerät zum abarbeiten eines computerprogramms | |
EP1149338B1 (de) | Lastverteilungsverfahren eines multiprozessorsystems und multiprozessorsystem | |
DE102016221526A1 (de) | Vorrichtung und Verfahren zum Bearbeiten einer Mehrzahl Aufgaben | |
EP1514180A2 (de) | Reaktionszeit-beschränkung eines software-prozesses | |
DE102004059972A1 (de) | Thread-Scheduling-Verfahren, und Einrichtung zur Verwendung bei einem Thread-Scheduling-Verfahren | |
DE10110444A1 (de) | Verfahren und Vorrichtung zum Ermitteln der Auslastung eines Rechengeräts | |
WO2018091329A1 (de) | Vorrichtung und verfahren zum bearbeiten von aufträgen | |
WO2006045733A2 (de) | Verfahren, betriebssystem und rechengerät zum abarbeiten eines computerprogramms | |
DE102016113968A1 (de) | Prozessor zur korrelationsbasierter Endlosschleifenerkennung | |
WO2021249616A1 (de) | Verfahren zum konfigurieren von komponenten in einem system mit hilfe von multi-agent reinforcement learning, computerlesbares speichermedium und system | |
DE102022200160A1 (de) | Verfahren zum Optimieren eines Prozesses | |
EP1502189B1 (de) | Verfahren zur ermittlung der prioritätsabhangigen rechenzeitverteilung in einem priorit tsgesteuerten mehrprozess-rechenysystem | |
WO2009034063A2 (de) | Edf-implementierung für realzeitsysteme mit statischen prioritäten | |
DE10018660B4 (de) | Verfahren und Vorrichtung zum Steuern eines oder mehrerer elektromechanischer Stellantriebe | |
EP3690563A1 (de) | Verteilung einer steueraufgabe auf mehrere prozessorkerne | |
WO2007048675A1 (de) | Verfahren zum implementieren von software-timern und datenverarbeitungssystem | |
DE102021101309A1 (de) | Elektronische steuervorrichtung | |
EP1889159A2 (de) | Verfahren zur speicherverwaltung von digitalen recheneinrichtungen |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2006723199 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007557434 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
NENP | Non-entry into the national phase |
Ref country code: RU |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: RU |
|
WWP | Wipo information: published in national office |
Ref document number: 2006723199 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11884916 Country of ref document: US |