CN103942101A - Real-time task scheduling method and system based on multicore network processor - Google Patents
Real-time task scheduling method and system based on multicore network processor Download PDFInfo
- Publication number
- CN103942101A CN103942101A CN201310022689.3A CN201310022689A CN103942101A CN 103942101 A CN103942101 A CN 103942101A CN 201310022689 A CN201310022689 A CN 201310022689A CN 103942101 A CN103942101 A CN 103942101A
- Authority
- CN
- China
- Prior art keywords
- task
- message
- module
- processing
- time
- 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.)
- Pending
Links
Abstract
The invention provides a real-time task scheduling method and system based on a multicore network processor. The real-time task scheduling system comprises a message receiving module, a task scheduling module, a task processing module and a message transmitting module, wherein the message receiving module is used for acquiring messages from the multicore network processor actively and transmitting the acquired messages to the task scheduling module; the task scheduling module is used for storing the received messages into a multistage work array and preserving a task status machine for each message; the task processing module is used for processing the messages in the multistage work array in a polling mode; the message transmitting module is used for transmitting the processed messages out. The real-time task scheduling method and system based on the multicore network processor has the advantages that on a multicore network processor platform, various hardware resources in the multicore network processor are made full use of and integrated, a high-efficiency task scheduling mechanism is achieved, and processing performance of task processing software in the multicore network processor is improved.
Description
Technical field
The present invention relates to the method for scheduling task based on multi-core network processor platform, be specifically related to a kind of real-time task scheduling method and system based on multi-core network processor.
Background technology
Along with the continuous progress of chip manufacturing process, traditional single core processor architecture technique has faced bottleneck, it is more than one hundred million that transistorized integrated level has exceeded, be difficult to merely carry out improving performance by improving dominant frequency, and the raising of the dominant frequency raising that simultaneously also brings power consumption.So polycaryon processor arises at the historic moment, by increasing the quantity of concurrent physical processor in computing machine, can effectively utilize the concurrency of thread, realize executed in parallel truly.
Chip multi-core processor (CMP) is that multiple calculating kernels are integrated in sheet processor chips, thereby improves computing power.Each microcontroller core is in fact a relatively simple single-threading microprocessor or fairly simple multithreaded microprocessor, just executive routine code concurrently of multi-microprocessor core like this, thereby there is higher Thread level parallelism.Because CMP has adopted relatively simple microprocessor as processor core, make that CMP has that high primary frequency, design and proving period are short, steering logic simple, favorable expandability, the advantage such as be easy to realize, low in energy consumption, communication delay is low.
, according to the equity of CPU core whether CMP can be divided into isomorphism multinuclear and heterogeneous polynuclear.Calculating kernel is identical, status equity be called isomorphism multinuclear.Isomorphism CMP great majority are made up of general processor, and multiple processors are carried out identical or similar task.Calculating kernel difference, what status was not reciprocity is called heterogeneous polynuclear, and heterogeneous polynuclear adopts main processing core+association to process the design of core conventionally.
Multi-core network processor, as Cavium company, RMI Corp. provide multi-core network processor.The universal CPU of the multinuclear of these processors based on MIPS64 framework.The inner integrated hardware-accelerated processing unit of L3-L7 of OCTEON family chip.
The challenge that network polycaryon processor software systems realize is that the hardware configuration relation of Software for Design and network processing unit is very tight, the hardware architecture programming of necessary Network Processor Oriented, the various hardware resources that are processing data packets by reasonable distribution and use network processing unit, for example, on multiprocessing engine, specialized hardware processing unit, all kinds of register, sheet internal memory and other hardware cell, just can obtain a high performance system.
So, need a kind of real-time software framework, make full use of and integrate the various hardware resources on polycaryon processor, realize a kind of Task Scheduling Mechanism efficiently, thereby improve the performance of the task process software on polycaryon processor.
Summary of the invention
The object of the invention is to, for overcoming the problems referred to above, the present invention proposes a kind of real-time task scheduling method and system based on multi-core network processor.
For achieving the above object, the invention provides a kind of real-time task scheduling system based on multi-core network processor, described system comprises: message sink module, task scheduling modules, task processing module and message transmission module;
Described message sink module, for initiatively obtaining message from multi-core network processor, and by the transmission of messages of obtaining to task scheduling modules;
Described task scheduling modules, for depositing the message receiving in multistage operating queue and being a task status machine of each message maintenance;
Described task processing module, for processing the message of multistage operating queue according to the mode of poll;
Described message transmission module, for sending message after treatment.
Above-mentioned task scheduling modules further comprises:
Encapsulation submodule, is packaged into task for the message of message sink module being obtained according to priority;
Task status machine maintenance update submodule, be used to the task of each encapsulation to safeguard a task status machine, and the mode that adopts state machine to shift completes the real-time follow-up of the treatment state to each task;
Output sub-module, exports the multistage operating queue with different priorities to for the processing of task processing module for the task that encapsulation submodule is packaged into;
Wherein, the task status of described task status machine record comprises: queuing armed state, at treatment state, process error status, queue up state to be sent and the state of finishing dealing with.
Above-mentioned task processing module further comprises:
Obtain the submodule of task, obtain task for the priority authority according to self from described multistage operating queue, and task is resolved;
Second processes submodule, for the task type providing according to task, task is processed accordingly;
Wherein, described task type comprises: system control task, protocol data task and Timing Processing task.
Also provide a kind of real-time task scheduling method based on multi-core network processor based on said system the present invention, the method is based on system claimed in claim 1, and described method comprises:
Step 101) message sink module initiatively obtains message from polycaryon processor, enters next step in the time getting message;
Step 102) task scheduling modules will obtain message according to priority and be packaged into task, and deposit this task in multistage operating queue according to priority, be that task that priority is high deposits the work queue of higher priority in and task low priority deposited in to the work queue of lower priority, the state machine of this task enters queuing armed state;
Step 103) task processing module is according to the queue of priority orders poll multistage operating, and from multistage operating queue, obtain and resolve task, and according to the task type providing in task, each task to be processed accordingly, this task enters treatment state;
Step 104) when task enters after the state of finishing dealing with, task processing module is packaged into corresponding message by task and packs transmit queue into, sends this message by message transmission module; In the time that task processing makes a mistake, return to step 103) until this task process successfully.
Step 103) described respective handling further comprises:
Step 103-1) in the time comprising system control task in task, task processing module is carried out corresponding system control action;
Step 103-2) in the time comprising protocol data task in task, task processing module is carried out task after the protocol processes of this layer, data is submitted to the CPU core processing of upper layer application;
Step 103-3) in the time comprising Timing Processing task in task, task processing module is mounted to task in the Timer timeslice of requirement;
Step 103-4) at step 103-1), 103-2) and 103-3) in, in to the processing procedure of task, make a mistake, by this task transfers to process error status, get back to step 102);
Step 103-5) at step 103-1), 103-2) and 103-3) in, when the processing success to task, by task transfers to processing completion status.
Above-mentioned message sink module is obtained message from all CPU cores, system timer or the network port of polycaryon processor on one's own initiative.
Above-mentioned message transmission module requires to send its message of obtaining according to different queue priorities and timer time sheet.
Compared with prior art, the invention has the beneficial effects as follows: on multi-core network processing platform, make full use of and integrated the various hardware resources on polycaryon processor, realized a kind of efficient Task Scheduling Mechanism, improved the handling property of the task process software on polycaryon processor.
Brief description of the drawings
Fig. 1 is the composition frame chart of the real-time task scheduling system based on multi-core network processor provided by the invention;
Fig. 2 is the process flow diagram of the real-time task scheduling method based on multi-core network processor provided by the invention;
Fig. 3 is the software framework structure schematic diagram that the embodiment of the present invention provides;
Fig. 4 is the concrete workflow diagram that the embodiment of the present invention provides.
Embodiment
Below in conjunction with drawings and Examples, the method for the invention is elaborated.
The present invention proposes a kind of real-time task scheduling system, this system makes full use of and integrates the various hardware resources on polycaryon processor, realizes a kind of Task Scheduling Mechanism efficiently, thereby improves the performance of the task process software on polycaryon processor.
The present invention is achieved through the following technical solutions:
By the Task Scheduling Mechanism of message-driven, on polycaryon processor, realize efficient task treatment scheme.This software frame comprises message sink module, task scheduling modules, task processing module and message transmission module.The message trigger that the workflow of this software frame is got by message sink module, task scheduling modules deposits the message receiving in multistage operating queue and is a task status machine of each message maintenance, task processing module is processed the message in multistage operating queue according to the mode of poll, finally by message transmission module, message after treatment is sent.
Message sink module is obtained message from other CPU cores, system timer (Timer) or the network port etc. on polycaryon processor on one's own initiative, and by the follow-up work flow process of this software frame of message trigger getting.
The message that task scheduling modules is obtained message sink module according to priority is packaged into task (WORK), and deposits the multistage operating queue with different priorities in.Simultaneously, task scheduling modules will be that each WORK safeguards a task status machine, mainly comprise queuing armed state, at treatment state, process error status, queue up state to be sent and the state of finishing dealing with, the mode that in system, the processing of each WORK is shifted by state machine completes.
Task processing module is obtained WORK according to the priority authority of self from multistage operating queue, WORK is resolved, according to the task type providing in WORK, WORK is processed accordingly, main task type comprises: system control task, protocol data task and Timing Processing task.
The workflow that this software frame is realized the task scheduling of message-driven is:
Step 1: at initial phase, message sink module is obtained message from other CPU cores, Timer or the network port etc. of polycaryon processor on one's own initiative, by the follow-up work flow process of this software frame of message trigger getting.
Step 2: the message that task scheduling modules is obtained message sink module according to priority is packaged into WORK, and deposit multistage operating queue according to priority, the state machine of this WORK enters queuing armed state;
Step 3: task processing module obtains and resolve WORK according to the priority authority of self from responsive operation queue, according to the task type providing in WORK, this WORK is processed accordingly, and this WORK enters at treatment state.
Step 3-1: in the time comprising system control task in WORK, task processing module is carried out corresponding system control action;
Step 3-2: in the time comprising protocol data task in WORK, task processing module is carried out WORK after the protocol processes of this layer, data is submitted to the CPU core processing of upper layer application;
Step 3-3: in the time comprising Timing Processing task in WORK, task processing module is mounted to task in the Timer timeslice of requirement;
Step 3-4: in step 3-1,3-2,3-3, make a mistake in to the processing procedure of WORK, WORK is transferred to processing error status, get back to step 2.
Step 3-5: in step 3-1,3-2,3-3, when the processing success to WORK, WORK is transferred to the state of finishing dealing with.
Step 4: when WORK enters after the state of finishing dealing with, task processing module is packaged into corresponding message by WORK and packs transmit queue into, message transmission module requires to send this message according to different queue priorities and Timer timeslice.Finally proceed to step 1 and continue to obtain message pending in system.
Embodiment
As shown in Figure 3, this software frame comprises message sink module 301, task scheduling modules 309, task processing module 303 and message transmission module 308 etc. to the software framework structure of embodiment.The message trigger that the workflow of this software frame is got by message sink module 301, task scheduling modules 309 deposits the message receiving in multistage operating queue 302 and is a task status machine 310 of each message maintenance, task processing module 303 is processed the message in multistage operating queue 302 according to the mode of poll, finally by message transmission module 308, message after treatment is sent.
As shown in Figure 4, software frame of the present invention workflow is in the present embodiment the task scheduling workflow of embodiment:
Step 401,402, after whole system initialization, message sink module 301 is obtained message from other CPU cores, Timer or the network port etc. of polycaryon processor on one's own initiative.
Step 403, when getting after a piece of news 311, the follow-up work flow process of this software frame is triggered and starts working.
Step 404, the message 311 that task scheduling modules 309 is obtained message sink module 301 according to priority is packaged into task (WORK), and deposits multistage operating queue 302 according to priority, and the task status machine 310 of this WORK enters queuing armed state.
Step 405, task processing module 303 obtains and resolves WORK according to the priority authority of self from multistage operating queue 302, and this WORK enters at treatment state.The type that obtains this WORK by parsing is iSCSI protocol processes task, by task processing module 303, the iSCSI PDU comprising in WORK is carried out to corresponding protocol processes.
Step 406, if there is mistake in task processes, the task status machine 310 of this WORK enters processing error status, gets back to afterwards step 405; If task is processed successfully, the task status machine 310 of WORK is transferred to the state of finishing dealing with, enter step 408.
Step 408, when WORK is in the state of finishing dealing with, task processing module 303 is packaged into corresponding message 312 by WORK and packs message send queue 307 into, and WORK enters the state to be sent of queuing up afterwards.
Step 409,410, message transmission module 308 requires to send message 312 according to the queue priority of message 312 and Timer timeslice.If sent unsuccessfully, get back to step 409 and resend; If sent successfully, this software frame is finished dealing with to message 311, and workflow is got back to step 401.
In a word, a kind of real-time task scheduling method provided by the invention and system, relate to the method for scheduling task based on multi-core network processor platform.The present invention proposes a kind of real-time task scheduling method, realize efficient task treatment scheme by the Task Scheduling Mechanism of message-driven.The method comprises message sink module, task scheduling modules, task processing module and message transmission module.The message trigger that the workflow of the method is got by message sink module, task scheduling modules deposits the message receiving in multistage operating queue and is a task status machine of each message maintenance, task processing module is processed the message in multistage operating queue according to the mode of poll, finally by message transmission module, message after treatment is sent.The hardware resource of network processing unit of the present invention, realizes a kind of Task Scheduling Mechanism efficiently, thereby improves the processing power of the task process software of multi-core network processor.
It should be noted last that, above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is modified or is equal to replacement, do not depart from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of claim scope of the present invention.
Claims (7)
1. the real-time task scheduling system based on multi-core network processor, described system comprises: message sink module, task scheduling modules, task processing module and message transmission module;
Described message sink module, for initiatively obtaining message from multi-core network processor, and by the transmission of messages of obtaining to task scheduling modules;
Described task scheduling modules, for depositing the message receiving in multistage operating queue and being a task status machine of each message maintenance;
Described task processing module, for processing the message of multistage operating queue according to the mode of poll;
Described message transmission module, for sending message after treatment.
2. the real-time task scheduling system based on multi-core network processor according to claim 1, is characterized in that, described task scheduling modules further comprises:
Encapsulation submodule, is packaged into task for the message of message sink module being obtained according to priority;
Task status machine maintenance update submodule, be used to the task of each encapsulation to safeguard a task status machine, and the mode that adopts state machine to shift completes the real-time follow-up of the treatment state to each task;
Output sub-module, exports the multistage operating queue with different priorities to for the processing of task processing module for the task that encapsulation submodule is packaged into;
Wherein, the task status of described task status machine record comprises: queuing armed state, at treatment state, process error status, queue up state to be sent and the state of finishing dealing with.
3. the real-time task scheduling system based on multi-core network processor according to claim 2, is characterized in that, described task processing module further comprises:
Obtain the submodule of task, obtain task for the priority authority according to self from described multistage operating queue, and task is resolved;
Second processes submodule, for the task type providing according to task, task is processed accordingly;
Wherein, described task type comprises: system control task, protocol data task and Timing Processing task.
4. the real-time task scheduling method based on multi-core network processor, the method is based on system claimed in claim 1, and described method comprises:
Step 101) message sink module initiatively obtains message from polycaryon processor, enters next step in the time getting message;
Step 102) task scheduling modules will obtain message according to priority and be packaged into task, and deposit this task in multistage operating queue according to priority, be that task that priority is high deposits the work queue of higher priority in and task low priority deposited in to the work queue of lower priority, the state machine of this task enters queuing armed state;
Step 103) task processing module is according to the queue of priority orders poll multistage operating, and from multistage operating queue, obtain and resolve task, and according to the task type providing in task, each task to be processed accordingly, this task enters treatment state;
Step 104) when task enters after the state of finishing dealing with, task processing module is packaged into corresponding message by task and packs transmit queue into, sends this message by message transmission module; In the time that task processing makes a mistake, return to step 103) until this task process successfully.
5. the real-time task scheduling method based on multi-core network processor according to claim 4, is characterized in that step 103) described respective handling further comprises:
Step 103-1) in the time comprising system control task in task, task processing module is carried out corresponding system control action;
Step 103-2) in the time comprising protocol data task in task, task processing module is carried out task after the protocol processes of this layer, data is submitted to the CPU core processing of upper layer application;
Step 103-3) in the time comprising Timing Processing task in task, task processing module is mounted to task in the Timer timeslice of requirement;
Step 103-4) at step 103-1), 103-2) and 103-3) in, in to the processing procedure of task, make a mistake, by this task transfers to process error status, get back to step 102);
Step 103-5) at step 103-1), 103-2) and 103-3) in, when the processing success to task, by task transfers to processing completion status.
6. the real-time task scheduling method based on multi-core network processor according to claim 4, is characterized in that, described message sink module is obtained message from all CPU cores, system timer or the network port of polycaryon processor on one's own initiative.
7. the real-time task scheduling method based on multi-core network processor according to claim 4, is characterized in that, described message transmission module requires to send its message of obtaining according to different queue priorities and timer time sheet.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310022689.3A CN103942101A (en) | 2013-01-21 | 2013-01-21 | Real-time task scheduling method and system based on multicore network processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310022689.3A CN103942101A (en) | 2013-01-21 | 2013-01-21 | Real-time task scheduling method and system based on multicore network processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103942101A true CN103942101A (en) | 2014-07-23 |
Family
ID=51189776
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310022689.3A Pending CN103942101A (en) | 2013-01-21 | 2013-01-21 | Real-time task scheduling method and system based on multicore network processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103942101A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016082463A1 (en) * | 2014-11-24 | 2016-06-02 | 中兴通讯股份有限公司 | Data processing method and apparatus for multi-core processor, and storage medium |
| CN108228327A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of task |
| CN108958904A (en) * | 2017-05-25 | 2018-12-07 | 北京忆恒创源科技有限公司 | The driver frame of the lightweight operating system of embedded multi-core central processing unit |
| CN109710399A (en) * | 2018-12-14 | 2019-05-03 | 湖北航天技术研究院总体设计所 | A kind of DSP communication task scheduling system and method |
| CN111209112A (en) * | 2019-12-31 | 2020-05-29 | 杭州迪普科技股份有限公司 | Exception handling method and device |
| CN111538604A (en) * | 2020-04-22 | 2020-08-14 | 北京字节跳动网络技术有限公司 | Distributed task processing system |
| CN112860389A (en) * | 2019-11-27 | 2021-05-28 | 上海哔哩哔哩科技有限公司 | Task scheduling method based on state machine |
| CN114780227A (en) * | 2022-06-20 | 2022-07-22 | 中国人民解放军国防科技大学 | Task scheduling mapping method and system based on core granular network processor architecture |
| CN116197919A (en) * | 2023-05-06 | 2023-06-02 | 曜石机器人(上海)有限公司 | Robot control system and control method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070101316A1 (en) * | 2005-09-12 | 2007-05-03 | Don Berndt | Software Architecture and Program for the Concurrent Execution of Finite State Machine-Encoded Processes, on Single or Multiple-Processor-Based Embedded Systems |
| US20100162253A1 (en) * | 2008-12-18 | 2010-06-24 | Samsung Electronics Co., Ltd. | Real-time scheduling method and central processing unit based on the same |
| CN102222022A (en) * | 2011-07-15 | 2011-10-19 | 范示德汽车技术（上海）有限公司 | Real-time task scheduling method based on multicore processor |
| CN102713852A (en) * | 2012-02-01 | 2012-10-03 | 华为技术有限公司 | Multicore processor system |
-
2013
- 2013-01-21 CN CN201310022689.3A patent/CN103942101A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070101316A1 (en) * | 2005-09-12 | 2007-05-03 | Don Berndt | Software Architecture and Program for the Concurrent Execution of Finite State Machine-Encoded Processes, on Single or Multiple-Processor-Based Embedded Systems |
| US20100162253A1 (en) * | 2008-12-18 | 2010-06-24 | Samsung Electronics Co., Ltd. | Real-time scheduling method and central processing unit based on the same |
| CN102222022A (en) * | 2011-07-15 | 2011-10-19 | 范示德汽车技术（上海）有限公司 | Real-time task scheduling method based on multicore processor |
| CN102713852A (en) * | 2012-02-01 | 2012-10-03 | 华为技术有限公司 | Multicore processor system |
Non-Patent Citations (1)
| Title |
|---|
| 郭秀岩: "面向多核的多层次实时网络数据流调度技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016082463A1 (en) * | 2014-11-24 | 2016-06-02 | 中兴通讯股份有限公司 | Data processing method and apparatus for multi-core processor, and storage medium |
| CN108958904A (en) * | 2017-05-25 | 2018-12-07 | 北京忆恒创源科技有限公司 | The driver frame of the lightweight operating system of embedded multi-core central processing unit |
| CN108958904B (en) * | 2017-05-25 | 2024-04-05 | 北京忆恒创源科技股份有限公司 | Driver framework of lightweight operating system of embedded multi-core central processing unit |
| CN108228327B (en) * | 2017-12-29 | 2021-01-15 | 北京奇虎科技有限公司 | Task processing method and device |
| CN108228327A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of task |
| CN109710399A (en) * | 2018-12-14 | 2019-05-03 | 湖北航天技术研究院总体设计所 | A kind of DSP communication task scheduling system and method |
| CN112860389A (en) * | 2019-11-27 | 2021-05-28 | 上海哔哩哔哩科技有限公司 | Task scheduling method based on state machine |
| CN111209112A (en) * | 2019-12-31 | 2020-05-29 | 杭州迪普科技股份有限公司 | Exception handling method and device |
| CN111538604A (en) * | 2020-04-22 | 2020-08-14 | 北京字节跳动网络技术有限公司 | Distributed task processing system |
| CN111538604B (en) * | 2020-04-22 | 2023-03-28 | 北京字节跳动网络技术有限公司 | Distributed task processing system |
| CN114780227A (en) * | 2022-06-20 | 2022-07-22 | 中国人民解放军国防科技大学 | Task scheduling mapping method and system based on core granular network processor architecture |
| CN116197919A (en) * | 2023-05-06 | 2023-06-02 | 曜石机器人(上海)有限公司 | Robot control system and control method |
| CN116197919B (en) * | 2023-05-06 | 2023-07-25 | 曜石机器人(上海)有限公司 | Robot control system and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103942101A (en) | Real-time task scheduling method and system based on multicore network processor | |
| US11836524B2 (en) | Memory interface for a multi-threaded, self-scheduling reconfigurable computing fabric | |
| US20220188265A1 (en) | Loop Thread Order Execution Control of a Multi-Threaded, Self-Scheduling Reconfigurable Computing Fabric | |
| Liu et al. | Design and evaluation of an rdma-aware data shuffling operator for parallel database systems | |
| Fusco et al. | High speed network traffic analysis with commodity multi-core systems | |
| US20190303147A1 (en) | Execution Control of a Multi-Threaded, Self-Scheduling Reconfigurable Computing Fabric | |
| Frey et al. | A spinning join that does not get dizzy | |
| US20190303327A1 (en) | Multi-Threaded, Self-Scheduling Reconfigurable Computing Fabric | |
| CN111078436B (en) | Data processing method, device, equipment and storage medium | |
| CN105187327A (en) | Distributed message queue middleware | |
| US20120297216A1 (en) | Dynamically selecting active polling or timed waits | |
| CN112035388B (en) | High-performance encryption and decryption method based on PCI-e channel | |
| US20210334228A1 (en) | Managing network interface controller-generated interrupts | |
| CN107562685B (en) | Method for data interaction between multi-core processor cores based on delay compensation | |
| CN104683472A (en) | Data transmission method capable of supporting large data volume | |
| CN102722402A (en) | Processing system and method for dispatching multiple tasks in real time in non-preemptive mode | |
| Sun et al. | Republic: Data multicast meets hybrid rack-level interconnections in data center | |
| CN113407479A (en) | Many-core architecture embedded with FPGA and data processing method thereof | |
| CN103916316A (en) | Linear speed capturing method of network data packages | |
| CN109918335A (en) | One kind being based on 8 road DSM IA frame serverPC system of CPU+FPGA and processing method | |
| CN115878301A (en) | Acceleration framework, acceleration method and equipment for database network load performance | |
| US11392388B2 (en) | System and method for dynamic determination of a number of parallel threads for a request | |
| CN110119375B (en) | Control method for linking multiple scalar cores into single-core vector processing array | |
| CN105099645A (en) | Multi-user concurrent communication method and device based on half-duplex communication device | |
| CN102929714A (en) | uC/OS-II-based hardware task manager |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140723 |
|
| RJ01 | Rejection of invention patent application after publication |