CN111880933A - Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform - Google Patents
Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform Download PDFInfo
- Publication number
- CN111880933A CN111880933A CN202010731008.0A CN202010731008A CN111880933A CN 111880933 A CN111880933 A CN 111880933A CN 202010731008 A CN202010731008 A CN 202010731008A CN 111880933 A CN111880933 A CN 111880933A
- Authority
- CN
- China
- Prior art keywords
- task
- hardware
- hardware task
- record table
- reconfigurable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Design And Manufacture Of Integrated Circuits (AREA)
Abstract
The invention discloses a reconfigurable hardware task dynamic allocation method based on a heterogeneous computing platform, which relates to the technical field of computers, and is characterized in that the placement position of a newly arrived hardware task on a reconfigurable chip is obtained through a rectangular vertex, the earliest latest start time priority principle is adopted, the real-time performance of the hardware task is improved, the resource recovery task of the completed hardware task is executed by adopting the idea of configuration reuse, the rejection rate of hardware task allocation is reduced, the utilization rate of the reconfigurable chip is improved, the expense of the computing platform is reduced, a two-dimensional reconfigurable hardware task is supported, and the scheduling efficiency is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a reconfigurable hardware task dynamic allocation method based on a heterogeneous computing platform.
Background
In a computing platform applied in the aerospace field, hardware tasks to be distributed cover traditional real-time control type hardware tasks, integer computation intensive hardware tasks, floating point computation intensive hardware tasks, communication intensive hardware tasks, I/O intensive hardware tasks and combinations of various types of hardware tasks. Different types of hardware tasks have greatly different demands on computing platforms.
The simplest and direct reconfigurable hardware task dynamic allocation method based on the heterogeneous computing platform is a queuing-based first-come first-serve method, namely, when a new hardware task arrives, whether available resources exist is checked, if yes, the hardware task is accepted, and if not, the hardware task is rejected. However, the distribution quality of the method is not good, and a more ideal distribution method should satisfy: as long as there is a blank configurable space in the reconfigurable device or there is enough capacity before the latest start time of the task, the allocation method should be able to search and guarantee the resource requirements of the subsequent hardware task to the maximum extent.
The existing dynamic allocation technical means for the hardware tasks on the reconfigurable chip mainly adopts a module gap minimization method, and the method utilizes a two-dimensional space-time coordinate system to express the relationship between tasks and time. The vertical axis represents time for executing the task, the horizontal axis represents space occupied by the task, and the whole distribution process is a process in which the task continuously extends upward. Selecting an optimal layout position and starting time for a newly arrived hardware task in two steps: finding all possible allocations that meet deadline constraints for newly arrived hardware tasks; among all possible allocations, the best one is selected. The main disadvantages of this method are as follows:
(1) the condition of configuration reuse is not considered, and the overhead of a computing platform is increased under the condition of more periodic tasks;
(2) the allocation is based on slack time, one important assumption being that the execution time of the hardware task is fixed and predictable, otherwise the allocation method degrades to a non-real-time allocation. In an actual computing scene, a hardware task is greatly different from a software task, the hardware task often repeatedly executes a data-driven task to complete a large number of data processing tasks, and the requirement on flexibility is not high. Because the data to be processed is usually related to the environment, the execution time of the hardware task cannot be accurately obtained in advance, and the rejection rate of the hardware task allocation is high;
(3) when the method is expanded to a two-dimensional reconfigurable hardware task, the complexity and the cost of the method are obviously increased, and the distribution efficiency is low.
Disclosure of Invention
In order to solve the defects of the prior art, the embodiment of the invention provides a reconfigurable hardware task dynamic allocation method based on a heterogeneous computing platform, which comprises the following steps:
s1, judging whether the hardware task Ti is the hardware task which is just executed, if yes, deleting the hardware task Ti from the activity record table A and inserting the hardware task Ti into the completed task record table F, meanwhile, updating the rectangular vertex layout record table C, and going to the step S2;
s2, taking out a hardware task Tj from the waiting record table W, judging whether a rectangle capable of accommodating the hardware task Tj exists in the completed task record table F, if so, moving the hardware task Tj from the completed task record table F to the active record table A, updating a rectangle vertex layout record table C, activating the hardware task Tj, and if not, turning to the step S3;
s3 judging whether the current time exceeds the latest start time of the hardware task Tj, if yes, refusing to dispatch the hardware task Tj and deleting the hardware task Tj, if not, going to step S4;
s4 judging whether the hardware task Tj can be distributed in the free rectangle vertex recording table B, if yes, using the dynamic reconfigurable resource layout method to lay the hardware task Tj in the free rectangle vertex recording table B and going to the step S6, if not, going to the step S5;
s5, according to the earliest and latest priority principle, finding the replaceable hardware task Tk of the hardware task Tj from the completed task record table F, deleting the hardware task Tk from the completed task record table F, updating the idle rectangular vertex record table B, and by using a dynamic reconfigurable resource layout method, laying out the hardware task Tj into the idle rectangular vertex record table B and turning to the step S6;
s6, inserting the hardware task Tj into the activity record table A, updating the rectangular vertex layout record table C, and activating the hardware task Tj;
s7 repeats steps S1-S6 until all hardware tasks in the waiting record W have been scanned.
Preferably, when the replaceable hardware task Tk of the hardware task Tj cannot be found in the completed task record table F, it is determined whether the hardware task Tj is distributable in the rectangular vertex layout record table C, if so, the hardware task Tj is laid out in the rectangular vertex layout record table C by using a dynamic reconfigurable resource layout method, the hardware task overlapping with the hardware task Tj in the completed task record table F is deleted, the free rectangular vertex record table B is updated, and the process proceeds to step S6.
Preferably, the determining whether the hardware task Tj can be allocated in the rectangular vertex layout record table C includes:
if not, the hardware task Tj is notified to wait and repeat steps S2-S6.
Preferably, the searching for the alternative hardware task Tk of the hardware task Tj from the completed task record table F according to the earliest-latest priority principle includes:
and searching a rectangle which is completed earliest and can contain the hardware task Tj from the completed task recording table F, taking the hardware task on the rectangle as a replaceable hardware task Tk of the hardware task Tj, replacing the hardware task Tk with the hardware task Tj, and updating the rectangular vertex layout recording table C.
Preferably, the determining whether the hardware task Ti is a hardware task that has just ended includes:
if not, the hardware tasks Ti are sequentially inserted into the waiting record table W according to the sequence of the latest start time, and the steps S2-S6 are repeated.
The method for dynamically allocating the reconfigurable hardware tasks based on the heterogeneous computing platform has the following beneficial effects:
the method comprises the steps of obtaining the placement position of a newly arrived hardware task on a reconfigurable chip through a rectangular vertex, adopting the earliest latest start time priority principle, improving the real-time performance of the hardware task, adopting the idea of configuration reuse to execute the resource recovery task of the completed hardware task, reducing the rejection rate of hardware task allocation, improving the utilization rate of the reconfigurable chip, reducing the expense of a computing platform, supporting the two-dimensional reconfigurable hardware task, and having high scheduling efficiency.
Drawings
Fig. 1 is a flowchart illustrating a method for dynamically allocating reconfigurable hardware tasks based on a heterogeneous computing platform according to an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and the embodiments.
Interpretation of terms
Dynamic reconfiguration: the dynamic reconfiguration is also called as runtime reconfiguration, which means that different configuration information is input or exchanged for reconfigurable hardware during system runtime, and the structure of the hardware is changed in real time to adapt to different hardware layouts, so that the chip utilization rate and the system operation efficiency are higher;
the reconfigurable chip comprises: is composed of a certain number of minimum reconfigurable logic units and interconnection logic among the minimum reconfigurable logic units;
the reconfigurable logic unit: the reconfigurable chip is a minimum unit of reconfigurable chip configuration, and when circuit functions are changed arbitrarily and slightly, a system at least needs a minimum unit of configuration;
hardware tasks: generally, a functional module which can be configured into a reconfigurable device after synthesis is referred to, and is generally a rectangular block;
the dynamic reconfigurable resource layout method comprises the following steps: the method can lay out the new hardware tasks in a matrix form according to the resource distribution condition of the current reconfigurable chip, so that the resources of the reconfigurable chip can be fully utilized.
The method for dynamically allocating the reconfigurable hardware tasks based on the heterogeneous computing platform comprises the following steps:
s101, judging whether the hardware task Ti is a hardware task which is just executed, if so, deleting the hardware task Ti from the activity record table A and inserting the hardware task Ti into the completed task record table F, updating the rectangular vertex layout record table C, and turning to the step S102.
Wherein, the top point in the completed task record table F cannot be reallocated and can only be reused. Therefore, after a space is found in the completed task record table F, the space is released to be in an idle state, and then the space is deallocated to the idle rectangle vertex record table B.
S102, taking out a hardware task Tj from the waiting record table W, judging whether a rectangle capable of accommodating the hardware task Tj exists in the completed task record table F, if so, moving the hardware task Tj from the completed task record table F to the active record table A, updating a rectangular vertex layout record table C, activating the hardware task Tj, and if not, turning to the step S103.
S103, judging whether the current time exceeds the latest starting time of the hardware task Tj, if so, refusing to schedule the hardware task Tj and deleting the hardware task Tj, and if not, turning to the step S104.
If the hardware task Tj cannot be allocated before the latest start time, the real-time property of the hardware task Tj cannot be guaranteed, and therefore the hardware task is rejected and cannot be scheduled.
S104, judging whether the hardware task Tj can be distributed in the idle rectangular vertex recording table B, if so, utilizing a dynamic reconfigurable resource layout method to lay the hardware task Tj in the idle rectangular vertex recording table B and turning to the step S106, and if not, turning to the step S105.
The purpose of this step is to retain executed tasks on the reconfigurable device as much as possible and to increase the allocation hit rate.
S105, according to the earliest and latest priority principle, searching the replaceable hardware task Tk of the hardware task Tj from the completed task recording table F, deleting the hardware task Tk from the completed task recording table F, updating the idle rectangular vertex recording table B, and distributing the hardware task Tj into the idle rectangular vertex recording table B by using a dynamic reconfigurable resource distribution method and transferring to the step S106.
S106, inserting the hardware task Tj into the activity record table A, updating the rectangular vertex layout record table C, and activating the hardware task Tj.
S107 repeats steps S101-S106 until all hardware tasks in the waiting log W have been scanned.
Optionally, when the replaceable hardware task Tk of the hardware task Tj cannot be found in the completed task record table F, it is determined whether the hardware task Tj is distributable in the rectangular vertex layout record table C, if so, the hardware task Tj is laid out in the rectangular vertex layout record table C by using a dynamic reconfigurable resource layout method, the hardware task overlapping with the hardware task Tj in the completed task record table F is deleted, the free rectangular vertex record table B is updated, and the process goes to step S106.
Optionally, the determining whether the hardware task Tj can be allocated in the rectangular vertex layout record table C includes:
if not, the hardware task Tj is notified to wait and repeat steps S102-S106.
Optionally, the searching for the alternative hardware task Tk of the hardware task Tj from the completed task record table F according to the earliest-latest priority principle includes:
and searching a rectangle which is completed earliest and can contain the hardware task Tj from the completed task recording table F, taking the hardware task on the rectangle as a replaceable hardware task Tk of the hardware task Tj, replacing the hardware task Tk with the hardware task Tj, and updating the rectangular vertex layout recording table C.
The vertices in the rectangular vertex layout record table C are not considered for reuse when distributing the hardware tasks, and any hardware task is deleted from the rectangular vertex layout record table C after being executed.
Optionally, the determining whether the hardware task Ti is a hardware task that has just ended includes:
if not, the hardware tasks Ti are sequentially inserted into the waiting record table W according to the sequence of the latest start time, and the steps S102-S106 are repeated.
According to the reconfigurable hardware task dynamic allocation method based on the heterogeneous computing platform, the placement position of a newly arrived hardware task on the reconfigurable chip is obtained through the rectangular vertex, the earliest latest starting time priority principle is adopted, the real-time performance of the hardware task is improved, the resource recovery task of the completed hardware task is executed by adopting the configuration reuse idea, the rejection rate of hardware task allocation is reduced, the utilization rate of the reconfigurable chip is improved, the expense of the computing platform is reduced, the two-dimensional reconfigurable hardware task is supported, and the scheduling efficiency is improved.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
It will be appreciated that the relevant features of the method and apparatus described above are referred to one another. In addition, "first", "second", and the like in the above embodiments are for distinguishing the embodiments, and do not represent merits of the embodiments.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (5)
1. A reconfigurable hardware task dynamic allocation method based on a heterogeneous computing platform is characterized by comprising the following steps:
s1, judging whether the hardware task Ti is the hardware task which is just executed, if yes, deleting the hardware task Ti from the activity record table A and inserting the hardware task Ti into the completed task record table F, meanwhile, updating the rectangular vertex layout record table C, and going to the step S2;
s2, taking out a hardware task Tj from the waiting record table W, judging whether a rectangle capable of accommodating the hardware task Tj exists in the completed task record table F, if so, moving the hardware task Tj from the completed task record table F to the active record table A, updating a rectangle vertex layout record table C, activating the hardware task Tj, and if not, turning to the step S3;
s3 judging whether the current time exceeds the latest start time of the hardware task Tj, if yes, refusing to dispatch the hardware task Tj and deleting the hardware task Tj, if not, going to step S4;
s4 judging whether the hardware task Tj can be distributed in the free rectangle vertex recording table B, if yes, using the dynamic reconfigurable resource layout method to lay the hardware task Tj in the free rectangle vertex recording table B and going to the step S6, if not, going to the step S5;
s5, according to the earliest and latest priority principle, finding the replaceable hardware task Tk of the hardware task Tj from the completed task record table F, deleting the hardware task Tk from the completed task record table F, updating the idle rectangular vertex record table B, and by using a dynamic reconfigurable resource layout method, laying out the hardware task Tj into the idle rectangular vertex record table B and turning to the step S6;
s6, inserting the hardware task Tj into the activity record table A, updating the rectangular vertex layout record table C, and activating the hardware task Tj;
s7 repeats steps S1-S6 until all hardware tasks in the waiting record W have been scanned.
2. The method for dynamically allocating reconfigurable hardware tasks based on heterogeneous computing platforms according to claim 1 is characterized in that when the replaceable hardware task Tk of the hardware task Tj cannot be found in the completed task record table F, whether the hardware task Tj can be allocated in the rectangular vertex layout record table C is judged, if yes, the hardware task Tj is laid out in the rectangular vertex layout record table C by using a dynamic reconfigurable resource layout method, the hardware task Tj in the completed task record table F, which overlaps with the hardware task Tj, is deleted, the free rectangular vertex record table B is updated, and the process proceeds to step S6.
3. The method for dynamically allocating the reconfigurable hardware tasks based on the heterogeneous computing platform according to claim 2, wherein the step of judging whether the hardware task Tj can be allocated in the rectangular vertex layout record table C comprises the following steps:
if not, the hardware task Tj is notified to wait and repeat steps S2-S6.
4. The method for dynamically allocating reconfigurable hardware tasks based on heterogeneous computing platforms according to claim 1, wherein the step of searching the alternative hardware task Tk of the hardware task Tj from the completed task record table F according to the earliest-latest priority principle comprises the following steps:
and searching a rectangle which is completed earliest and can contain the hardware task Tj from the completed task recording table F, taking the hardware task on the rectangle as a replaceable hardware task Tk of the hardware task Tj, replacing the hardware task Tk with the hardware task Tj, and updating the rectangular vertex layout recording table C.
5. The method for dynamically allocating the reconfigurable hardware tasks based on the heterogeneous computing platform according to claim 1, wherein the step of judging whether the hardware task Ti is the hardware task which is just finished comprises the following steps:
if not, the hardware tasks Ti are sequentially inserted into the waiting record table W according to the sequence of the latest start time, and the steps S2-S6 are repeated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010731008.0A CN111880933B (en) | 2020-07-27 | 2020-07-27 | Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010731008.0A CN111880933B (en) | 2020-07-27 | 2020-07-27 | Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111880933A true CN111880933A (en) | 2020-11-03 |
| CN111880933B CN111880933B (en) | 2023-09-22 |
Family
ID=73200728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010731008.0A Active CN111880933B (en) | 2020-07-27 | 2020-07-27 | Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111880933B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6415384B1 (en) * | 1998-10-30 | 2002-07-02 | Lucent Technologies Inc. | Hardware/software co-synthesis of dynamically reconfigurable embedded systems |
| EP1372084A2 (en) * | 2002-05-31 | 2003-12-17 | Interuniversitair Microelektronica Centrum Vzw | Method for hardware-software multitasking on a reconfigurable computing platform |
| CN101799770A (en) * | 2010-01-19 | 2010-08-11 | 湖南大学 | Reconfigurable resource management method based on unit area speed-up ratio |
| CN101833368A (en) * | 2010-04-13 | 2010-09-15 | 杭州电子科技大学 | Method for managing energy of local dynamic reconfigurable system by coordinatively processing software and hardware |
| CN102681901A (en) * | 2012-05-08 | 2012-09-19 | 西安交通大学 | Segmental reconfigurable hardware task arranging method |
| CN103488531A (en) * | 2013-09-26 | 2014-01-01 | 中国船舶重工集团公司第七一六研究所 | Software and hardware mixing real-time task scheduling method based on multi-core processor and FPGA |
| CN104239135A (en) * | 2014-05-19 | 2014-12-24 | 江苏科技大学 | Hybrid task scheduling method of directed acyclic graph (DGA) based reconfigurable system |
| CN104615488A (en) * | 2015-01-16 | 2015-05-13 | 华为技术有限公司 | Task scheduling method and device on heterogeneous multi-core reconfigurable computing platform |
| US20160202999A1 (en) * | 2013-08-19 | 2016-07-14 | Topic Embedded Systems B.V. | A Computing Platform, A Reconfigurable Hardware Device And A Method for Simultaneously Executing Processes On Dynamically Reconfigurable Hardware Device, Such As An FPGA, As Well As Instruction Set Processors, Such As A CPU, And A Related Computer Readable Medium |
| CN108256182A (en) * | 2018-01-02 | 2018-07-06 | 西安交通大学 | A kind of layout method of dynamic reconfigurable FPGA |
-
2020
- 2020-07-27 CN CN202010731008.0A patent/CN111880933B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6415384B1 (en) * | 1998-10-30 | 2002-07-02 | Lucent Technologies Inc. | Hardware/software co-synthesis of dynamically reconfigurable embedded systems |
| EP1372084A2 (en) * | 2002-05-31 | 2003-12-17 | Interuniversitair Microelektronica Centrum Vzw | Method for hardware-software multitasking on a reconfigurable computing platform |
| CN101799770A (en) * | 2010-01-19 | 2010-08-11 | 湖南大学 | Reconfigurable resource management method based on unit area speed-up ratio |
| CN101833368A (en) * | 2010-04-13 | 2010-09-15 | 杭州电子科技大学 | Method for managing energy of local dynamic reconfigurable system by coordinatively processing software and hardware |
| CN102681901A (en) * | 2012-05-08 | 2012-09-19 | 西安交通大学 | Segmental reconfigurable hardware task arranging method |
| US20160202999A1 (en) * | 2013-08-19 | 2016-07-14 | Topic Embedded Systems B.V. | A Computing Platform, A Reconfigurable Hardware Device And A Method for Simultaneously Executing Processes On Dynamically Reconfigurable Hardware Device, Such As An FPGA, As Well As Instruction Set Processors, Such As A CPU, And A Related Computer Readable Medium |
| CN103488531A (en) * | 2013-09-26 | 2014-01-01 | 中国船舶重工集团公司第七一六研究所 | Software and hardware mixing real-time task scheduling method based on multi-core processor and FPGA |
| CN104239135A (en) * | 2014-05-19 | 2014-12-24 | 江苏科技大学 | Hybrid task scheduling method of directed acyclic graph (DGA) based reconfigurable system |
| CN104615488A (en) * | 2015-01-16 | 2015-05-13 | 华为技术有限公司 | Task scheduling method and device on heterogeneous multi-core reconfigurable computing platform |
| CN108256182A (en) * | 2018-01-02 | 2018-07-06 | 西安交通大学 | A kind of layout method of dynamic reconfigurable FPGA |
Non-Patent Citations (2)
| Title |
|---|
| R. FITCH ET AL: ""Reconfiguration planning for heterogeneous self-reconfiguring robots"", 《 PROCEEDINGS 2003 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS 2003)》 * |
| 刘彦;李仁发;许新达;徐成;: "一种异构可重构片上系统的实时任务调度算法", 计算机研究与发展, no. 06 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111880933B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110647394B (en) | Resource allocation method, device and equipment | |
| CN110489213B (en) | Task processing method and processing device and computer system | |
| CN100383745C (en) | Facilitating allocation of resources in a heterogeneous computing environment | |
| US8700838B2 (en) | Allocating heaps in NUMA systems | |
| US8650570B2 (en) | Method of assigning instructions in a process to a plurality of scheduler instances based on the instruction, in which each scheduler instance is allocated a set of negoitaited processor resources | |
| KR101626378B1 (en) | Apparatus and Method for parallel processing in consideration of degree of parallelism | |
| CN114741207B (en) | GPU resource scheduling method and system based on multi-dimensional combination parallelism | |
| US20040210900A1 (en) | Providing predictable scheduling of programs using repeating precomputed schedules on discretely scheduled and/or multiprocessor operating systems | |
| KR20090066765A (en) | Method and apparatus for migrating task in multicore platform | |
| CN102609467A (en) | Method and device for database connection management | |
| CN112905342A (en) | Resource scheduling method, device, equipment and computer readable storage medium | |
| CN105677744A (en) | Method and apparatus for increasing service quality in file system | |
| CN110084507B (en) | Scientific workflow scheduling optimization method based on hierarchical perception in cloud computing environment | |
| CN117435324B (en) | Task scheduling method based on containerization | |
| CN111880933B (en) | Reconfigurable hardware task dynamic allocation method based on heterogeneous computing platform | |
| CN113225269B (en) | Container-based workflow scheduling method, device and system and storage medium | |
| CN115964152A (en) | GPU resource scheduling method, equipment and storage medium | |
| CN115509704A (en) | Task scheduling method, device, equipment and storage medium | |
| CN111736998A (en) | Memory management method and related product | |
| CN111881639B (en) | Dynamic layout method for reconfigurable hardware task | |
| CN113301087A (en) | Resource scheduling method, device, computing equipment and medium | |
| Kaladevi et al. | Processor co-allocation enabling advanced reservation of jobs in MultiCluster systems | |
| CN116541018B (en) | Distributed model compiling system, method, device, medium and equipment | |
| CN111124594B (en) | Container operation method and device, heterogeneous GPU (graphics processing Unit) server and container cluster system | |
| KR20180076051A (en) | Method for processing OpenCL kernel and computing device thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |