CN102508712A - Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof - Google Patents
Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof Download PDFInfo
- Publication number
- CN102508712A CN102508712A CN2011103017392A CN201110301739A CN102508712A CN 102508712 A CN102508712 A CN 102508712A CN 2011103017392 A CN2011103017392 A CN 2011103017392A CN 201110301739 A CN201110301739 A CN 201110301739A CN 102508712 A CN102508712 A CN 102508712A
- Authority
- CN
- China
- Prior art keywords
- task
- layer
- application program
- hardware
- tasks
- 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
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a middleware system of a heterogeneous multi-core reconfigurable hybrid system and a task execution method thereof. The middleware system is characterized by comprising an application program management layer, a task dividing and scheduling layer, a driving and input/output layer and a communication interface layer, wherein the application program management layer is used for supplying the operation environment and programming interface of an application program to a user; the task dividing and scheduling layer is in charge of scheduling in the process of dividing, mapping and operating tasks; and the driving and input/output layer is used for integrating the drives as well as input and output interfaces of a plurality of IP (Internet Protocol) cores; the application program management layer is used for providing an application program interface and an operation run time library for an application layer; the task dividing and scheduling layer used for dividing and scheduling task requests through a task dividing module, a mapping module and a scheduling module; the driving and input/output layer is integrated with a hardware drive and utilizes a hardware resource of an FPGA (Field Programmable Gate Array) platform; and the communication interface layer is in charge of transmitting data between reconfigurable hardware platforms. According to the system provided by the invention, the throughput rate of the platform and the performance of the system are improved by carrying out relative monitoring and automatic parallelization on the tasks in the operating process. The system can be used for supplying middleware supports comprising task dividing, scheduling and the like to a heterogeneous reconfigurable hybrid system, so that the execution efficiency is improved.
Description
Technical field
The invention belongs to the middleware Technology field of chip multi-core computing platform, be specifically related to middleware system and task executing method in a kind of heterogeneous polynuclear restructural commingled system.
Background technology
Restructural (Run Time Reconfiguration) when dynamic reconfigurable is otherwise known as operation; Be meant and when operation, carry out reconstruct; Be that reconfigurable hardware injects or the different configuration information that swaps out when being the program execution; Change the structure of hardware in real time, adapting to different hardware algorithms, thereby have higher chip utilization factor and running efficiency of system.
At the end of the nineties in last century, successively occurred several kinds and be used for specific area, based on the dynamic reconfigurable system of customization reconfigurable hardware.PipeRench is a kind of streamlined coarseness reconfigurable system that CMU proposes, and its research and development start from 1997, and in flow in 2002 success.A task is divided into some flowing water stage, can makes the reconfigurable hardware operation surpass the virtual task of its storage size.Its shortcoming is that range of application is fixed, and each pipelining-stage can only be the function of a current state and a last pipelining-stage.Morphosys adopts the coarseness structure, and the degree of depth is able to programme, and Wonder is 32 about it, and shortcoming is the configuration underaction, can only be applied to some task of Flame Image Process.Similarly project also has Paddi, Cameron, and Chess Array, Rapid etc., they are all in extendability, and the versatility aspect runs into certain difficulty.
At present, except the restructural chip, comprise one or several general processors toward contact in the dynamic reconfigurable system, it is used to handle tasks such as resource management, reconstruct.Usually claim that this system is a heterogeneous polynuclear restructural commingled system.In this system, how hardware is carried out abstractly, and to the user access interface and programming model to be provided be to need at present the difficult problem that solves.
Generally speaking operating system is positioned at the upper strata of hardware, provides certain abstract, can simplify the exploitation of application program, and under operating system environment, using no longer is the individuality that isolates.Operating system provides with using independently various mutually services (as: fifo buffer, timer, semaphore, driver etc.) and supplies the user to call, and the task that the user submits to realizes using relevant specific function on a higher level.Reconfigurable hardware (especially part restructural FPGA) can be regarded the Heterogeneous Computing resource with ISP equality as; Through being its design operation system, can accelerate the development progress of the system of blending together, shorten Time To Market; Reduce cost of development; Enhanced system portability and dirigibility reduce design complexities and maintenance difficulties, can support effectively that more dynamic reconfigurable blendes together multitask, the multi-user environment in the system; Improve the utilization factor and the configuration speed of restructural chip, thereby reach the purpose of giving full play to computing power.Since the dynamic reconfigurable system occurs, especially, there have been many scholars to carry out exploratory research both at home and abroad in this field along with the widespread use of the system of blending together.
Just be suggested in recent years because restructural blendes together the notion of operating system, and also existed more problem needs further to explore and research at present.The present invention therefore.
Summary of the invention
In order to overcome the deficiency that exists in the background technology; One object of the present invention is to provide middleware and the task execution model in a kind of heterogeneous polynuclear restructural commingled system; It comprises an application program management layer; Task division and dispatch layer drive and the I/O layer, and the communication interface layer constitute.Middleware in the heterogeneous polynuclear restructural commingled system that the present invention proposes can provide middleware supports such as task division scheduling for the heterogeneous reconfigurable commingled system, thereby improves task executions efficient.
In order to solve these problems of the prior art, technical scheme provided by the invention is:
Middleware system in a kind of heterogeneous polynuclear restructural commingled system is characterized in that said middleware system comprises the task division of when the user provides the running environment of application program and the application program management layer of DLL, the division of being responsible for carrying out task, mapping and operation, dispatching and dispatch layer, the driving of integrated multiple IP kernel and driving and the I/O layer and the communication interface layer of input and output interface; Application programming interfaces and run-time library that said application program management layer provides application layer to need; Said task division and dispatch layer are divided scheduling through task division module, mapping block and scheduler module to task requests and are handled; Said driving and I/O layer integrated hardware drive the hardware resource that utilizes the FPGA platform; Said communication interface layer is responsible for carrying out the data transmission between the reconfigurable hardware platform.
Preferably, said communication interface layer comprises hardware interface, software interface and restructural interface; Hardware interface is used for the communication between microprocessor and the Hardware I P nuclear, and detects the interrupt request of interconnection through interruptable controller; Software interface is used for transmission information between two microprocessors; The restructural interface is used for the reconfiguring of bit stream of IP kernel.
Preferably, during the integrated operation of said application program management layer analysis module with monitoring and analysis application program.
The present invention also provides the task executing method of the middleware system in a kind of heterogeneous polynuclear restructural commingled system, it is characterized in that said method comprised with the next stage:
(1) variable allocated phase: to each need carrying out of task, obtain its input and output parametric variable, and it is saved in the storage inside buffering;
(2) the task division stage:, confirm the target computing unit of operation to each task;
(3) tasks execution phases: task is sent to task in the computing unit in the commingled system through interconnected nuclear communication mechanism;
(4) the tasks synchronization stage carries out synchronously after task is finished, and the result is returned the user.
Preferably, the operand of each task distributes a list item during said method variable allocated phase; If there is not present variable in this table, then variable will be by built-in variable of implicit expression rename.
Preferably, said method task division is divided task through following method during the stage: if there is idle Hardware I P nuclear, task is sent to hardware, otherwise task will be sent to general processor.
Preferably, through interrupting returning results of task execution, adopt the First come first served tactful handling interrupt requests of service earlier during said method tasks execution phases.
Preferably, said method tasks synchronization and presentation stage adopt the grid synchronization mechanism, and each table all comprises many computing unit clauses and subclauses, this table tracing task sum and the number of tasks of having accomplished; If both equate, then indicate the equal end of run of all tasks, then can carry out synchronously; When tasks synchronization, be primarily aimed at relevant inspection of data between task, and guarantee that all tasks all return in order.
Application program management course user provides the running environment and the DLL of application program, and its constituted mode mainly comprises application programming interfaces and run-time library.Dispatch when said task division and dispatch layer mainly are responsible for carrying out division, mapping and the operation of task, therefore comprise task division module, mapping block and scheduler module.The integrated driving and the input and output interface of multiple IP kernel in said driving and the I/O layer.Said communication interface layer comprises hardware interface, software interface and restructural interface.
Task execution model in the heterogeneous polynuclear restructural commingled system of the present invention comprises four-stage: variable allocated phase, task division stage, tasks execution phases and tasks synchronization stage constitute.Said variable allocated phase to each need carrying out of task, obtains its input and output parametric variable, and it is saved in the storage inside buffering, as the basis of follow-up work.In the said task division stage,, confirm the target computing unit of operation to each task.Said tasks execution phases sends to task in the computing unit in the commingled system with task through interconnected nuclear communication mechanism.In the said tasks synchronization stage, after task is finished, carry out synchronously, and the result is returned the user.
The present invention relates to a kind of heterogeneous polynuclear restructural commingled system, be specifically related to middleware and task execution model in a kind of heterogeneous polynuclear restructural commingled system.Middleware and task execution model in the heterogeneous polynuclear restructural commingled system that the present invention proposes can provide middleware supports such as task division scheduling for the heterogeneous reconfigurable commingled system, thereby improve task executions efficient.
Compared with prior art, the present invention has following significant advantage:
1, the heterogeneous polynuclear restructural commingled system of the present invention's proposition provides the abstract of heterogeneous polynuclear reconfigurable hardware, thereby to the restructural multiple nucleus system middleware support is provided.
2, the task execution model of the present invention's proposition can provide the operating system grade support of running environment and division and scheduling for task, thus the automatic division and the dynamic dispatching of realizing of task.
In sum, the invention provides middleware and task execution model in a kind of heterogeneous polynuclear restructural commingled system, it comprises an application program management layer, and task division and dispatch layer drive and the I/O layer, and the communication interface layer constitute.Said application program management layer comprises application programming interfaces and run-time library; Said task division and dispatch layer comprise task division module, mapping block and scheduler module; The integrated driving and the input and output interface of multiple IP kernel in said driving and the I/O layer, said communication interface layer comprises hardware interface, software interface and restructural interface.Task execution model involved in the present invention comprises four-stage: variable allocated phase, task division stage, tasks execution phases and tasks synchronization stage constitute.Middleware and task execution model in the heterogeneous polynuclear restructural commingled system that the present invention proposes can provide middleware supports such as task division scheduling for the heterogeneous reconfigurable commingled system, thereby improve task executions efficient.
Description of drawings
Below in conjunction with accompanying drawing and embodiment the present invention is further described:
Fig. 1 is the middleware level block diagram in the heterogeneous polynuclear restructural commingled system of the present invention;
Fig. 2 is the process flow diagram that task execution model involved in the present invention comprises four-stage;
Embodiment
Above-mentioned explanation only is the general introduction of technical scheme of the present invention, understands technological means of the present invention in order can more to know, and can implement according to the content of instructions, below with preferred embodiment of the present invention and conjunction with figs. specify as after.Below in conjunction with specific embodiment such scheme is further specified.Should be understood that these embodiment are used to the present invention is described and are not limited to limit scope of the present invention.
Embodiment
As shown in Figure 1; Be the middleware level block diagram in the heterogeneous polynuclear restructural commingled system; The present invention relates to middleware and task execution model in a kind of heterogeneous polynuclear restructural commingled system, it comprises an application program management layer, task division and dispatch layer; Drive and the I/O layer, and the communication interface layer constitutes.Said application program management layer comprises application programming interfaces and run-time library; Said task division and dispatch layer comprise task division module, mapping block and scheduler module; The integrated driving and the input and output interface of multiple IP kernel in said driving and the I/O layer, said communication interface layer comprises hardware interface, software interface and restructural interface.
(1) application program management layer
For a task executions environment is provided, mainly comprise user's run-time library.These user libraries and be to provide through API (API) with other kernel library.API shows a high-level abstract view, the API that inner realization is defined, and user interface will keep a persistence unit, even reconfigure hardware, this API page or leaf remains unchanged.
In addition, analysis module during integrated operation in the application program management layer is with the monitoring and the analysis technology of support application program.The application program management layer is supported the monitoring and the method for decomposing of traditional application program, comprises analysis of central issue, the internal memory inspection of no initializtion, and dynamic information flow is followed the tracks of array bounds checking and soft error checked operation.The monitoring method in these whens operation can improve the security and the reliability of system and row able to programme.In addition,, can obtain the high-frequency part instant heating point code of program run, can be used for instructing the IP configuration through analyzing the focus that obtains.
(2) task division and dispatch layer
Task division and dispatching method before task is sent to IP kernel, have determined task is moved on which functional unit.Middleware of the present invention is compared with operating system, provides the support (like Xlinx DPR or SR) from reconstruct, and this system that makes has the advantage that when operation, dynamically reconfigures.Static nucleus module is to realize separately with reconfiguring module, and wherein having only reconstructed module is when operation, to reconfigure, to reduce the time overhead that bit stream is downloaded.At task division and dispatch layer, it is integrated to reconfigure core library.After IP kernel reconfigured, task need be repartitioned and dispatch.Therefore, the mapping meeting of task and IP kernel is by real-time update.
For each IP kernel, the execution time of specific tasks, speed-up ratio, area and power consumption etc. is carried out task division by the user, with the load balancing of realization system and the handling capacity of Geng Gao.
Because FPGA is a platform that area is limited, different IP kernels are all in contention limited hardware resource.The fundamental purpose of task scheduling is to improve the platform handling capacity, and the service efficiency of FPGA.
(3) driver and I/O layer
In order to utilize the hardware resource that is integrated in the FPGA platform, this layer mainly is responsible for the driver of integrated peripheral hardware and accumulator system.Driver commonly used mainly comprises: the interconnection driver is to be used for allowing between the application program of microprocessor and IP kernel transmission data and control messages to carry out through bus or network-on-chip.File system driver is with visiting local file.Static bit stream and local configuration bit stream are stored in file system.After task is finished, the result will return a look-at-me, thereby interruptable controller is the main transmission of being responsible for a plurality of task return datas.UART serial ports and timer mainly are responsible for user's debugging.The local memory storage system introduction of each microprocessor.The dynamic restructuring that reconfigurable controller is responsible for IP kernel configuration during with operation.
(4) communication interface layer
The communication interface layer is responsible for carrying out the data transmission between the reconfigurable hardware platform.Platform provides 3 kinds of basic data types: unified software interface (USI), and unified hardware interface (UHI), and unified interface (URI): the USI that reconfigures is mainly used in transmission information between two microprocessors.Comprised serial built-in function among the USI.Communication between UHI main users microprocessor and the Hardware I P nuclear, and pass through the interrupt request that interruptable controller detects interconnection.URI only is used for the reconfiguring of bit stream of IP section.
Fig. 2 is the process flow diagram that task execution model involved in the present invention comprises four-stage.The reconstruct flow example of IP kernel is following during a concrete operation:
(1) allocated phase of variable
Task list is shown do_T_adder (* in, * out), and * in representes to import the start address of array, and * out representes to export array.Relevant in order to detect data automatically, the task execution model need be collected all operations number of each sending of task.But we can only keep limited operand, rather than unlimited user-defined variable (at most of programming models, any operand that the user can use them to want).Therefore, the operand of each task needs a list item to distribute.If there is not present variable in this table, then variable will be by built-in variable of implicit expression rename.
(2) the task division stage
Task is sent to before the computing unit, and the task execution model must determine the computing unit on the task run.We take following strategy in the present invention: if there is idle Hardware I P nuclear, task is sent to hardware, otherwise task will be sent to general processor.
(3) tasks execution phases
Computing unit can begin to carry out after receiving all operations number automatically.And through interrupting return results.The interrupt request of an interruptable controller testing result and updating task variable.In the task execution model, because can return at one time, so the strategy that adopts First come first served to serve earlier is to be used for handling interrupt from the possibility of result of various tasks.
(4) tasks synchronization and submission
In case the dispatch service health check-up measures all calculating and has accomplished the unit execution, then gets into synchronous phase.The task execution model utilizes a grid synchronization mechanism.Each table all comprises many computing unit clauses and subclauses.This table tracing task sum and the number of tasks of having accomplished.If both equate, then indicate the equal end of run of all tasks, then can carry out synchronously.When tasks synchronization, be primarily aimed at relevant inspection of data between task, and guarantee that all tasks all return in order.
Above-mentioned instance only is explanation technical conceive of the present invention and characteristics, and its purpose is to let the people who is familiar with this technology can understand content of the present invention and enforcement according to this, can not limit protection scope of the present invention with this.All equivalent transformations that spirit is done according to the present invention or modification all should be encompassed within protection scope of the present invention.
Claims (8)
1. the middleware system in the heterogeneous polynuclear restructural commingled system is characterized in that said middleware system comprises the task division of when the user provides the running environment of application program and the application program management layer of DLL, the division of being responsible for carrying out task, mapping and operation, dispatching and dispatch layer, the driving of integrated multiple IP kernel and driving and the I/O layer and the communication interface layer of input and output interface; Application programming interfaces and run-time library that said application program management layer provides application layer to need; Said task division and dispatch layer are divided scheduling through task division module, mapping block and scheduler module to task requests and are handled; Said driving and I/O layer integrated hardware drive the hardware resource that utilizes the FPGA platform; Said communication interface layer is responsible for carrying out the data transmission between the reconfigurable hardware platform.
2. the middleware system in the heterogeneous polynuclear restructural commingled system according to claim 1 is characterized in that said communication interface layer comprises hardware interface, software interface and restructural interface; Hardware interface is used for the communication between microprocessor and the Hardware I P nuclear, and detects the interrupt request of interconnection through interruptable controller; Software interface is used for transmission information between two microprocessors; The restructural interface is used for the reconfiguring of bit stream of IP kernel.
3. the middleware system in the heterogeneous polynuclear restructural commingled system according to claim 1, analysis module is with monitoring and analysis application program when it is characterized in that the integrated operation of said application program management layer.
4. the task executing method of the middleware system in the described heterogeneous polynuclear restructural of claim 1 commingled system is characterized in that said method comprised with the next stage:
(1) variable allocated phase: to each need carrying out of task, obtain its input and output parametric variable, and it is saved in the storage inside buffering;
(2) the task division stage:, confirm the target computing unit of operation to each task;
(3) tasks execution phases: task is sent to task in the computing unit in the commingled system through interconnected nuclear communication mechanism;
(4) the tasks synchronization stage carries out synchronously after task is finished, and the result is returned the user.
5. according to the method for claim 4, the operand of each task distributes a list item when it is characterized in that said method variable allocated phase; If there is not present variable in this table, then variable will be by built-in variable of implicit expression rename.
6. according to the method for claim 4, it is characterized in that said method task division divides task through following method during the stage:, task is sent to hardware, otherwise task will be sent to general processor if there is idle Hardware I P nuclear.
7. according to the method for claim 4, through interrupting returning results of task execution, adopt the First come first served tactful handling interrupt requests of service earlier when it is characterized in that said method tasks execution phases.
8. according to the method for claim 4, it is characterized in that said method tasks synchronization and presentation stage adopt the grid synchronization mechanism, each table all comprises many computing unit clauses and subclauses, this table tracing task sum and the number of tasks of having accomplished; If both equate, then indicate the equal end of run of all tasks, then can carry out synchronously; When tasks synchronization, be primarily aimed at relevant inspection of data between task, and guarantee that all tasks all return in order.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110301739.2A CN102508712B (en) | 2011-09-29 | 2011-09-29 | Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110301739.2A CN102508712B (en) | 2011-09-29 | 2011-09-29 | Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102508712A true CN102508712A (en) | 2012-06-20 |
| CN102508712B CN102508712B (en) | 2014-01-15 |
Family
ID=46220804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110301739.2A Active CN102508712B (en) | 2011-09-29 | 2011-09-29 | Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102508712B (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902654A (en) * | 2012-09-03 | 2013-01-30 | 东软集团股份有限公司 | Method and device for linking data among heterogeneous platforms |
| CN102929577A (en) * | 2012-10-08 | 2013-02-13 | 上海大学 | FPGA (Field Programmable Gata Array) task placement maintaining method based on block splitting |
| CN103684849A (en) * | 2013-11-22 | 2014-03-26 | 汉柏科技有限公司 | Method and system for shielding network devices |
| CN103677984A (en) * | 2012-09-20 | 2014-03-26 | 中国科学院计算技术研究所 | Internet of Things calculation task scheduling system and method |
| CN104050033A (en) * | 2013-03-15 | 2014-09-17 | 辉达公司 | System and method for hardware scheduling of indexed barriers |
| CN104956328A (en) * | 2014-01-28 | 2015-09-30 | 华为技术有限公司 | Reentrant resource scheduling method, device and system |
| US9195506B2 (en) | 2012-12-21 | 2015-11-24 | International Business Machines Corporation | Processor provisioning by a middleware processing system for a plurality of logical processor partitions |
| US9298508B2 (en) | 2012-12-21 | 2016-03-29 | International Business Machines Corporation | Processor provisioning by a middleware processing system |
| WO2016082320A1 (en) * | 2014-11-26 | 2016-06-02 | 浪潮(北京)电子信息产业有限公司 | Hybrid heterogeneous host system, resource configuration method and task scheduling method |
| WO2016155335A1 (en) * | 2015-03-27 | 2016-10-06 | 华为技术有限公司 | Task scheduling method and device on heterogeneous multi-core reconfigurable computing platform |
| CN107301145A (en) * | 2017-06-29 | 2017-10-27 | 中国电子科技集团公司第二十九研究所 | Publish/subscribe formula software bus based on RapidIO under FPGA environment |
| CN108491268A (en) * | 2018-03-15 | 2018-09-04 | 浙江大学 | A kind of mixed core operating system joining automobile towards intelligent network |
| CN110096374A (en) * | 2019-03-20 | 2019-08-06 | 西安电子科技大学 | Communication middleware control system and method between internal multi-class computing unit |
| CN110399221A (en) * | 2019-07-23 | 2019-11-01 | 江苏鼎速网络科技有限公司 | Data processing method, system and terminal device |
| CN110398921A (en) * | 2019-07-24 | 2019-11-01 | 理工雷科电子(西安)有限公司 | A kind of multitask dynamic restructuring processing system and method based on FPGA control |
| CN111813526A (en) * | 2020-07-10 | 2020-10-23 | 深圳致星科技有限公司 | Heterogeneous processing system, processor and task processing method for federal learning |
| CN112306500A (en) * | 2020-11-30 | 2021-02-02 | 上海交通大学 | Compiling method for reducing multi-class access conflict aiming at coarse-grained reconfigurable structure |
| CN112506087A (en) * | 2019-09-16 | 2021-03-16 | 阿里巴巴集团控股有限公司 | FPGA acceleration system and method, electronic device, and computer-readable storage medium |
| CN113032035A (en) * | 2019-12-24 | 2021-06-25 | Oppo广东移动通信有限公司 | Configuration file processing method and device and electronic equipment |
| WO2024002172A1 (en) * | 2022-06-29 | 2024-01-04 | 上海寒武纪信息科技有限公司 | System on chip, instruction system, compilation system, and related product |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129390A (en) * | 2011-03-10 | 2011-07-20 | 中国科学技术大学苏州研究院 | Task scheduling system of on-chip multi-core computing platform and method for task parallelization |
| CN102135950A (en) * | 2011-03-10 | 2011-07-27 | 中国科学技术大学苏州研究院 | On-chip heterogeneous multi-core system based on star type interconnection structure, and communication method thereof |
-
2011
- 2011-09-29 CN CN201110301739.2A patent/CN102508712B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129390A (en) * | 2011-03-10 | 2011-07-20 | 中国科学技术大学苏州研究院 | Task scheduling system of on-chip multi-core computing platform and method for task parallelization |
| CN102135950A (en) * | 2011-03-10 | 2011-07-27 | 中国科学技术大学苏州研究院 | On-chip heterogeneous multi-core system based on star type interconnection structure, and communication method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 王超: "异构多核可重构片上系统关键技术研究", 《中国博士学位论文全文数据库》 * |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9250986B2 (en) | 2012-09-03 | 2016-02-02 | Neusoft Corporation | Method and apparatus for data linkage between heterogeneous platforms |
| CN102902654A (en) * | 2012-09-03 | 2013-01-30 | 东软集团股份有限公司 | Method and device for linking data among heterogeneous platforms |
| CN102902654B (en) * | 2012-09-03 | 2015-08-26 | 东软集团股份有限公司 | Data linkage method and device between heterogeneous platform |
| CN103677984A (en) * | 2012-09-20 | 2014-03-26 | 中国科学院计算技术研究所 | Internet of Things calculation task scheduling system and method |
| CN103677984B (en) * | 2012-09-20 | 2016-12-21 | 中国科学院计算技术研究所 | A kind of Internet of Things calculates task scheduling system and method thereof |
| CN102929577A (en) * | 2012-10-08 | 2013-02-13 | 上海大学 | FPGA (Field Programmable Gata Array) task placement maintaining method based on block splitting |
| US9280393B2 (en) | 2012-12-21 | 2016-03-08 | International Business Machines Corporation | Processor provisioning by a middleware processing system for a plurality of logical processor partitions |
| US9195506B2 (en) | 2012-12-21 | 2015-11-24 | International Business Machines Corporation | Processor provisioning by a middleware processing system for a plurality of logical processor partitions |
| US9298508B2 (en) | 2012-12-21 | 2016-03-29 | International Business Machines Corporation | Processor provisioning by a middleware processing system |
| US9311149B2 (en) | 2012-12-21 | 2016-04-12 | International Business Machines Corporation | Processor provisioning by a middleware processing system |
| CN104050033A (en) * | 2013-03-15 | 2014-09-17 | 辉达公司 | System and method for hardware scheduling of indexed barriers |
| CN103684849A (en) * | 2013-11-22 | 2014-03-26 | 汉柏科技有限公司 | Method and system for shielding network devices |
| CN104956328A (en) * | 2014-01-28 | 2015-09-30 | 华为技术有限公司 | Reentrant resource scheduling method, device and system |
| CN104956328B (en) * | 2014-01-28 | 2018-12-14 | 华为技术有限公司 | One kind can reentry resource regulating method, equipment and system |
| US9904577B2 (en) | 2014-11-26 | 2018-02-27 | Inspur (Beijing) Electronic Information Industry Co., Ltd | Hybrid heterogeneous host system, resource configuration method and task scheduling method |
| WO2016082320A1 (en) * | 2014-11-26 | 2016-06-02 | 浪潮(北京)电子信息产业有限公司 | Hybrid heterogeneous host system, resource configuration method and task scheduling method |
| WO2016155335A1 (en) * | 2015-03-27 | 2016-10-06 | 华为技术有限公司 | Task scheduling method and device on heterogeneous multi-core reconfigurable computing platform |
| US10452605B2 (en) | 2015-03-27 | 2019-10-22 | Huawei Technologies Co., Ltd. | Method and apparatus for task scheduling on heterogeneous multi-core reconfigurable computing platform |
| CN107301145B (en) * | 2017-06-29 | 2019-11-19 | 中国电子科技集团公司第二十九研究所 | Publish/subscribe formula software bus under FPGA environment based on RapidIO |
| CN107301145A (en) * | 2017-06-29 | 2017-10-27 | 中国电子科技集团公司第二十九研究所 | Publish/subscribe formula software bus based on RapidIO under FPGA environment |
| CN108491268A (en) * | 2018-03-15 | 2018-09-04 | 浙江大学 | A kind of mixed core operating system joining automobile towards intelligent network |
| CN108491268B (en) * | 2018-03-15 | 2021-10-08 | 浙江大学 | Intelligent networking automobile-oriented mixed core operating system |
| CN110096374A (en) * | 2019-03-20 | 2019-08-06 | 西安电子科技大学 | Communication middleware control system and method between internal multi-class computing unit |
| CN110096374B (en) * | 2019-03-20 | 2023-03-14 | 西安电子科技大学 | System and method for controlling communication middleware between internal multi-class computing units |
| CN110399221A (en) * | 2019-07-23 | 2019-11-01 | 江苏鼎速网络科技有限公司 | Data processing method, system and terminal device |
| CN110398921A (en) * | 2019-07-24 | 2019-11-01 | 理工雷科电子(西安)有限公司 | A kind of multitask dynamic restructuring processing system and method based on FPGA control |
| CN112506087A (en) * | 2019-09-16 | 2021-03-16 | 阿里巴巴集团控股有限公司 | FPGA acceleration system and method, electronic device, and computer-readable storage medium |
| CN113032035A (en) * | 2019-12-24 | 2021-06-25 | Oppo广东移动通信有限公司 | Configuration file processing method and device and electronic equipment |
| WO2021129339A1 (en) * | 2019-12-24 | 2021-07-01 | Oppo广东移动通信有限公司 | Method and device for processing configuration file, electronic apparatus, and storage medium |
| CN113032035B (en) * | 2019-12-24 | 2022-07-12 | Oppo广东移动通信有限公司 | Configuration file processing method and device and electronic equipment |
| CN111813526A (en) * | 2020-07-10 | 2020-10-23 | 深圳致星科技有限公司 | Heterogeneous processing system, processor and task processing method for federal learning |
| CN112306500A (en) * | 2020-11-30 | 2021-02-02 | 上海交通大学 | Compiling method for reducing multi-class access conflict aiming at coarse-grained reconfigurable structure |
| WO2024002172A1 (en) * | 2022-06-29 | 2024-01-04 | 上海寒武纪信息科技有限公司 | System on chip, instruction system, compilation system, and related product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102508712B (en) | 2014-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102508712B (en) | Middleware system of heterogeneous multi-core reconfigurable hybrid system and task execution method thereof | |
| Anderson et al. | Treehouse: A case for carbon-aware datacenter software | |
| Grasso et al. | LibWater: heterogeneous distributed computing made easy | |
| CN102023844B (en) | Parallel processor and thread processing method thereof | |
| US9462048B2 (en) | System and method for program and resource allocation within a data-intensive computer | |
| CN106598731B (en) | Runtime system and its control method based on heterogeneous polynuclear framework | |
| CN104899089A (en) | Task scheduling method in heterogeneous multi-core architecture | |
| CN105359098A (en) | Dynamic reconfiguration of applications on a multi-processor embedded system | |
| Rigo et al. | Paving the way towards a highly energy-efficient and highly integrated compute node for the Exascale revolution: the ExaNoDe approach | |
| CN101777007A (en) | Parallel function simulation system for on-chip multi-core processor and method thereof | |
| Vaishnav et al. | Heterogeneous resource-elastic scheduling for CPU+ FPGA architectures | |
| CN101976204B (en) | Service-oriented heterogeneous multi-core computing platform and task scheduling method used by same | |
| CN110297661A (en) | Parallel computing method, system and medium based on AMP framework DSP operating system | |
| CN101976201A (en) | CPU affinity-based virtual CPU dynamic binding method | |
| Zhang et al. | Models of parallel computation: a survey and classification | |
| CN110502334A (en) | Bandwidth aware task stealing method, system and chip based on mixing memory architecture | |
| CN103049305B (en) | Multithreading method for the dynamic code conversion of Godson multi-core CPU simulation | |
| D'Angelo et al. | Time warp on the go (updated version) | |
| CN103299277A (en) | Gpu system and processing method thereof | |
| Han et al. | Multi-core architectures with dynamically reconfigurable array processors for the WIMAX physical layer | |
| Paulin | Programming challenges & solutions for multi-processor SoCs: an industrial perspective | |
| Garcia et al. | Dynamic Percolation: A case of study on the shortcomings of traditional optimization in Many-core Architectures | |
| Ventroux et al. | Scmp architecture: an asymmetric multiprocessor system-on-chip for dynamic applications | |
| CN108073389A (en) | A kind of automotive engine system based on script | |
| Li et al. | A BPF-Based Task Scheduling Scheme for Heterogeneous Multiprocessor Systems |
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 |