WO2012055248A1 - 信息投递的方法和装置 - Google Patents

信息投递的方法和装置 Download PDF

Info

Publication number
WO2012055248A1
WO2012055248A1 PCT/CN2011/075893 CN2011075893W WO2012055248A1 WO 2012055248 A1 WO2012055248 A1 WO 2012055248A1 CN 2011075893 W CN2011075893 W CN 2011075893W WO 2012055248 A1 WO2012055248 A1 WO 2012055248A1
Authority
WO
WIPO (PCT)
Prior art keywords
task
target task
core
kernel
information
Prior art date
Application number
PCT/CN2011/075893
Other languages
English (en)
French (fr)
Inventor
赵阳
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to EP11835522.1A priority Critical patent/EP2634694B1/en
Priority to US13/881,018 priority patent/US9207974B2/en
Priority to RU2013121783/08A priority patent/RU2566953C2/ru
Priority to BR112013010079-6A priority patent/BR112013010079A2/pt
Publication of WO2012055248A1 publication Critical patent/WO2012055248A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/48Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806Task transfer initiation or dispatching
    • G06F9/4843Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/505Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/54Interprogram communication
    • G06F9/544Buffers; Shared memory; Pipes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00Indexing scheme relating to G06F9/00
    • G06F2209/50Indexing scheme relating to G06F9/50
    • G06F2209/502Proximity

Definitions

  • the present invention relates to the field of multi-core technology, and in particular, to a method and apparatus for information delivery. Background technique
  • each processor consists of multiple cores, each containing one or more hardware threads.
  • the communication device due to the complexity of the service, it consists of multiple service modules, control modules, and switch boards. Each board uses one or more multi-core processors.
  • the information communication method in the existing multi-core system is shown in Figure 1.
  • the information communication method follows the traditional single-core system processing mode, and the tasks between boards and boards pass IPC (Inter-Process Communication). ) Ways to achieve information communication.
  • the IPC method implements information communication based on TCP/IP or shared memory.
  • this information communication method cannot reflect the advantages of a multi-core system.
  • the complexity of the mechanism for the core information communication unit is greatly increased.
  • the processing capability of the control plane is overloaded. The reliability is reduced.
  • the IPC method cannot implement load balancing for multi-core system processing, which makes it impossible to achieve efficient communication when multi-instance of tasks in a multi-core system. Summary of the invention
  • the present invention provides a method for information delivery, which includes the steps of: receiving a to-be-delivered information and a target task name sent by a source task;
  • the corresponding kernel is searched for in the multi-core system according to the kernel number, and the to-be-delivered information is sent to the task in the kernel corresponding to the target task name.
  • the preset global relationship table includes: a correspondence between a task name in the multi-core system and a kernel number of a kernel where the task is located.
  • the source task and the target task are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in the core of different multi-core systems.
  • the target task includes one or more target task instances; the plurality of target task instances are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in different multi-core systems. Inside.
  • the method for information delivery when the target task includes multiple target task instances, selects the target task instance as the target task to be delivered according to the principle of proximity.
  • the invention further provides an apparatus for information delivery, comprising:
  • a receiving module configured to receive the to-be-delivered information and the target task name sent by the source task
  • a searching module configured to search, in the preset global relationship table, a kernel number corresponding to the target task name according to the target task name
  • a sending module configured to search for a corresponding kernel in the multi-core system according to the kernel number, and to be preferentially, the preset global relationship table includes: a correspondence between a task name in the multi-core system and a kernel number of the kernel where the task is located.
  • the source task and the target task are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in the core of different multi-core systems.
  • the target task includes one or more target task instances; the plurality of target task instances are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in different multi-core systems.
  • the searching module is further configured to: when the target task includes multiple target task instances, select the target task instance as the target task to be delivered according to the proximity principle.
  • the method and device for information delivery provided by the present invention can realize the delivery of information between multi-core systems or multi-core systems, and the reliability of information delivery is high, and the speed and efficiency of information delivery are improved, and further, Load balancing of task instances in a multi-core system while preventing the failure of information delivery.
  • FIG. 1 is a schematic structural view of a multi-core system in the prior art
  • FIG. 2 is a schematic flow chart of an embodiment of a method for information delivery according to the present invention.
  • FIG. 3 is a schematic structural view of a multi-core system of the present invention.
  • Figure 4 is a schematic structural view of a multi-core system of the present invention.
  • FIG. 5 is a schematic structural diagram of an apparatus for information delivery according to the present invention.
  • Figure 6 is another schematic structural view of the multi-core system of the present invention.
  • Fig. 7 is a schematic view showing another structure between the multi-core systems of the present invention. detailed description
  • the method includes: Step 101: Receive a to-be-delivered information and a target task name sent by a source task.
  • Step 102 Search for a kernel number corresponding to the target task name in the preset global relationship table according to the target task name;
  • Step 103 Search for a corresponding kernel in the multi-core system according to the kernel number, and send the to-be-delivered information to the task in the kernel corresponding to the target task name.
  • the task corresponding to the target task name in the kernel is the target task.
  • the preset global relationship table includes: a correspondence between a task name in the multi-core system and a kernel number of a kernel where the task is located.
  • the above method for information delivery is applied in a multi-core system, and the relationship between the task name and the kernel number of the kernel in which the task is located refers to the correspondence between all task names in the multi-core system and the kernel number of the kernel in which the task name is located, or a communication connection. All task names in all multicore systems correspond to the kernel number of the kernel in which the task resides.
  • the source task and the target task are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in the core of different multi-core systems.
  • the target task includes one or more target task instances; the plurality of target task instances are included in the same core of the same multi-core system, or include different cores in the same multi-core system. Internal, or included in the core of a different multi-core system.
  • the multi-core system 1 is an independent multi-core system, and the multi-core system 1 includes: a core 1, a core 2, and a kernel 3.
  • the kernel 1 includes the task A-1 and the task C-1;
  • the kernel 2 includes the task A-2 and the task B-1;
  • the kernel 3 includes the task D-1 and the task E-1.
  • the preset global relation table includes the following relationship between the task name and the kernel number of the kernel where the task is located: A-l ⁇ 1, A-2 ⁇ 2, B-l ⁇ 2, C-l ⁇ 1, D-l ⁇ 3, E-l ⁇ 3.
  • the multi-core systems 1, 2, and 3 are multi-core systems that communicate with each other, wherein the multi-core system 1 includes a core i, and the core i includes tasks a-1 and b-1.
  • the multi-core system 2 includes a core ii and a core iii, the core ii includes a task a-2 and a task d-1; the core iii includes a task a-3 and a task c-1.
  • the multi-core system 3 includes a kernel iv and a core V, and the kernel iv includes a task b-2 and a task d-2; the kernel V includes a task e-1 and a task f-1.
  • the preset global relation table includes the following relationship between the task name and the kernel number of the kernel where the task is located: al ⁇ i , a-2 ⁇ ii , A-3 ⁇ iii, bl ⁇ i, b-2 ⁇ iv, cl ⁇ iii, dl ⁇ ii, d-2 ⁇ iv, el ⁇ v, fl ⁇ v. Further, in the method embodiment of the above information delivery, when the target task has only one target task instance, the to-be-delivered information is directly sent to the target task instance.
  • the target task instance is selected as the target task to be delivered according to the proximity principle.
  • the principle of the proximity is as follows: the target task instance in the same kernel as the source task is preferentially selected as the target task to be delivered; if there is no target task instance in the same kernel as the source task, the priority is selected to be in the same multi-core system as the source task.
  • the target task instance is used as the target task to be delivered; if there is no target task instance in the same core as the source task and in the same multi-core system as the source task, the target task instance in the different multi-core system from the source task is selected as the target to be delivered.
  • the target task of the information is used as the target task to be delivered.
  • the load conditions of multiple target task instances in the target task and the specific conditions of the abnormality may be selected.
  • a target task instance is used as the target task of the information to be delivered.
  • the task d-1 is the source task, and the task d-1 needs to send the information to be delivered to the task a.
  • the task a includes three task instances: al, a-2, and a-3. Where task a-2 is in kernel ii, task a-3 is in kernel iii, task a-1 is in kernel i; task d-1 is in kernel ,, according to the proximity principle described above, preference is given to source task d -1
  • the target task instance a-2 in the same kernel acts as the target task for the information to be posted. If the target task instance selected as the target task has an abnormality or a high load condition, select one of the other target task instances as the target task to be delivered according to the same principle.
  • the target task instance a-2 is abnormal or the load is high
  • the target task instance a-3 that is in the same multi-core system as the source task d-1 is preferentially selected as the target task to be delivered information
  • Mission the target task instance a-2 and If a-3 is abnormal or the load is high
  • task c-1 be the source task
  • task c-1 send the message to be delivered to task d.
  • task d includes two task instances d-1 and d-2; source task c-1 and task d-1 are in the same multi-core system 2; task d-2 is in multi-core system 3; according to the principle of proximity described above
  • the task d-1 is preferentially selected as the target task of the information to be delivered; if the task d-1 is abnormal or the load is high, the task d-2 is selected as the target task of the information to be delivered.
  • task d-2 be the source task
  • task d-2 send the message to be delivered to task a.
  • task a includes three task instances: a-l, a-2, and a-3. All task instances a1, a-2, a-3 of task a are not in the same multi-core system as source task d-2; task a-1 is in multi-core system 1, and tasks a-2 and a-3 are in multi-core Within system 2. According to the above principle, the neighboring principle cannot be selected. In this case, it is necessary to select whether the three task instances are abnormal or the load is high or low, and the task instance with normal operation and low load is used as the target task of the information to be delivered.
  • the target task instance with the lowest load may be directly searched according to the principle of proximity, and the target task instance is used as the target task of the information to be delivered.
  • task a-1 be the source task
  • task b includes two task instances b-1 and b-2
  • task b-1 is heavily loaded.
  • the task b-2 has a low load, so the task b-2 is selected as the target task to be delivered, and the to-be-delivered information is sent to the task b-2 in the kernel iv. If an abnormal situation occurs in the target task instance selected as the target task, select one of the other target task instances as the target task to be delivered according to the same principle to ensure load balancing of different task instances in the same task, and also It can avoid the phenomenon of information delivery failure.
  • the method embodiment of the above information delivery can realize the inter-multi-core system or the multi-core system
  • an apparatus 10 for an information delivery apparatus which includes: a receiving module 11, a searching module 12, and a transmitting module 13.
  • the receiving module 11 is configured to receive the to-be-delivered information and the target task name sent by the source task
  • the searching module 12 is configured to search, in the preset global relationship table, the kernel number corresponding to the target task name according to the target task name; 13. For searching for a corresponding kernel in the multi-core system according to the kernel number, and sending the to-be-delivered information to the task in the kernel corresponding to the target task name.
  • the task corresponding to the target task name in the kernel is the target task.
  • the preset global relationship table includes: a correspondence between a task name in the multi-core system and a kernel number of a kernel where the task is located.
  • the information delivery device 10 is applied to the multi-core system, and the task name corresponds to the kernel number of the kernel where the task is located, and refers to the correspondence between all task names in the multi-core system and the kernel number of the kernel where the task name is located, or a communication connection. All task names in all multi-core systems correspond to the kernel number of the kernel in which the task resides.
  • the source task and the target task are included in the same core of the same multi-core system, or included in different cores of the same multi-core system, or included in a kernel of a different multi-core system.
  • the target task includes one or more target task instances; the plurality of target task instances are included in the same core of the same multi-core system, or include different cores in the same multi-core system. Internal, or included in the core of a different multi-core system.
  • the multi-core system 1 is an independent multi-core system equipped with the information delivery apparatus 10 of the present invention.
  • the multi-core system 1 includes: a core 1, a kernel 2, and a kernel 3.
  • kernel 1 includes task A-1 and task C-1;
  • kernel 2 includes task A-2 and task B-1;
  • Kernel 3 includes task D-1 and task El.
  • the preset global relationship table includes the following relationship between the task name and the kernel number of the kernel where the task is located: Al ⁇ 1, A-2 ⁇ 2, Bl ⁇ 2, Cl ⁇ 1, Dl ⁇ 3, El ⁇ 3.
  • the information to be delivered and the target task name are first sent to the device 10 for information delivery, and the device 10 for information delivery determines the name corresponding to the target task in the preset global relationship table according to the target task name.
  • the kernel number is then found in the multi-core system 1 with the kernel corresponding to the kernel number, and the to-be-delivered information is sent to the task in the kernel corresponding to the target task name.
  • the multi-core systems 1, 2, and 3 are multi-core systems that communicate with each other, and each of the multi-core systems is configured with the information delivery device 10 of the present invention.
  • the multi-core system 1 includes a kernel i, and the kernel i includes a task a-1 and a task b-1.
  • the multi-core system 2 includes a core ii and a core iii, the core ii includes a task a-2 and a task d-1; the core iii includes a task a-3 and a task c-1.
  • the multi-core system 3 includes a kernel iv and a kernel V, the kernel iv includes a task b-2 and a task d-2; the kernel v includes a task e-1 and a task f-l.
  • the preset global relationship table includes the following relationship between the task name and the kernel number of the kernel where the task is located: al ⁇ i, a-2 ⁇ ii, a-3 ⁇ iii, bl ⁇ i, b-2 ⁇ iv, cl ⁇ Iii, dl ⁇ ii, d-2 ⁇ iv, el ⁇ v, fl ⁇ v.
  • the information to be delivered and the target task name are first sent to the device 10 for information delivery, and the device 10 for information delivery determines the target in the preset global relationship table according to the target task name.
  • the kernel number corresponding to the task name finds the kernel corresponding to the kernel number in the multi-core systems 1, 2, and 3, and sends the to-be-delivered information to the task in the kernel corresponding to the target task name.
  • the to-be-delivered information is directly sent to the target task instance.
  • the searching module 12 is further configured to: when the target task includes multiple target task instances, select the target task instance as the target task to be delivered according to the principle of proximity.
  • the principle of the proximity is as follows: The target task instance in the same kernel as the source task is preferentially selected as the target task to be delivered; if not in the source task.
  • the target task instance of the same kernel preferentially selects the target task instance in the same multi-core system as the target task as the target task to be delivered; if there is no target task in the same kernel as the source task, and no target task in the same multi-core system as the source task
  • the target task instance that is in a different multi-core system from the source task is selected as the target task to be delivered information.
  • the device 10 for information delivery may according to the load condition and abnormality of multiple target task instances in the target task.
  • the specific case selects a target task instance as the target task of the information to be delivered.
  • the task d-1 is the source task, and the task d-1 needs to send the information to be delivered to the task a.
  • the task a includes three task instances: al, a-2, and a-3.
  • the task d-1 sends the to-be-delivered information and the target task name a to the information delivery device 10, and the information delivery device 10 searches the preset global relationship table for the kernel number corresponding to the target task name a according to the target task name a. There are i, ii, and iii (including the kernel number of the target task a).
  • the cores i, ii, and iii corresponding to the kernel numbers i, ii, and iii are then found in the multi-core systems 1, 2, and 3.
  • Task a-2 is in kernel ii
  • task a-3 is in kernel iii
  • task a-1 is in kernel i
  • task d-1 is in kernel ii, according to the proximity principle described above, information delivery device 10 is preferred
  • the source task d-1 is in the target task instance a-2 of the same kernel as the target task of the information to be delivered, and then the information to be delivered is sent to the task a-2 in the kernel ii.
  • the target task instance a-2 selected as the target task has an abnormality or a high load condition, select one of the other target task instances as the target task to be delivered according to the same principle. That is, if the target task instance a-2 is abnormal or the load is high, the information delivery device 10 preferentially selects the target task instance a-3 that is in the same multi-core system as the source task d-1 as the target task to be delivered, and then waits The delivery information is sent to task a-3 in kernel iii; if both target task instances a-2 and a-3 are abnormal or have a high load, then information delivery device 10 selects a target task instance that is in a different multi-core system than the source task.
  • Ai is the target task of the information to be delivered, and then the information to be delivered is sent to task a-1 in kernel i. That is, if selected as the target
  • the target task instance of the task has an abnormal condition and a high load condition, and in addition to the selected target task instance, there are two or more target task instances, and then one of the remaining target task instances is selected as the information to be delivered according to the same principle.
  • Target tasks to improve the speed and efficiency of information delivery, while avoiding information delivery failures.
  • task c-1 is the source task
  • task c-1 needs to send the information to be delivered to task d
  • task d includes two task instances d-1 and d-2.
  • the task c-1 sends the to-be-delivered information and the target task name d to the information delivery device 10, and the information delivery device 10 searches the preset global relationship table for the kernel number corresponding to the target task name d according to the target task name d.
  • ii and iv including the kernel number of the target task d).
  • the cores ii and iv corresponding to the core numbers ii and iv are then found in the multi-core systems 1, 2 and 3.
  • Source task c-1 is in kernel iii
  • task d-1 is in kernel ii
  • kernel ii and kernel iii are in the same multicore system 2
  • task d-2 is in kernel iv within multicore system 3; according to the proximity principle described above, information
  • the delivery device 10 preferentially selects the task d-1 as the target task to be delivered, and then transmits the to-be-delivered information to the task d-1; if the task d-1 is abnormal or the load is high, the information delivery device 10 selects the task d -2 is the target task of the information to be delivered, and then the information to be delivered is sent to task d-2.
  • task d-2 is the source task, and the task d-2 needs to send the information to be delivered to the task a.
  • task a includes three task instances: al, a-2, and a-3.
  • the task d-2 sends the to-be-delivered information and the target task name a to the information delivery device 10, and the information delivery device 10 searches for the kernel number corresponding to the target task name a in the preset global relationship table according to the target task name a.
  • the target task name a There are i, ii, and iii (including the kernel number of the target task a).
  • the cores i, ii, and iii corresponding to the kernel numbers i, ii, and iii are then found in the multi-core systems 1, 2, and 3. All instances a, a-2, a-3 of task a are not in the same core as source task d-2, nor are they in the same multicore system; task a-1 is in multicore system 1, and task a-2 And a-3 are in the multi-core system 2.
  • the device 10 for information delivery cannot select according to the principle of the above-mentioned proximity. At this time, it is necessary to select according to whether the three task instances of the task a are abnormal or the load is high or low, and the task that is normally operated and has a low load is required.
  • the instance acts as the target task for the information to be delivered.
  • the device for information delivery 10 can select the target task by the above-mentioned proximity principle, can improve the speed and efficiency of information delivery, avoid the phenomenon of information delivery failure, and ensure the load balancing of different task instances in the same task.
  • the target task instance with the lowest load may be directly searched according to the principle of proximity, and the target task instance is used as the target task of the information to be delivered.
  • the task a-1 is the source task, and the task a-1 needs to send the information to be delivered to the task d.
  • the task a-1 sends the to-be-delivered information and the target task name d to the device 10 for information delivery.
  • the information delivery device 10 searches the preset global relationship table for the kernel numbers corresponding to the target task name d according to the target task name d, ii and iv (including the kernel number of the target task d).
  • the kernels ii and iv corresponding to the kernel numbers ii, iv, and the task instances d-1 and d-2 are task instances, wherein the task d-1 is in the kernel ii, the task D-2 is in the kernel iv, the task d-1 is heavy, and the task d-2 is low, so the task d-2 of the kernel iv is selected as the target task to be delivered information, and the information delivery device 10 sends the information to be delivered. Go to task d-2 in kernel iv. If an abnormal situation occurs in the target task instance d-2 selected as the target task, the information delivery device sends the to-be-delivered information to task d-1 of kernel ii. If an abnormal situation occurs in the target task instance selected as the target task, and one of the other target task instances is selected as the target task to be delivered, to ensure load balancing of different task instances in the same task, and information can be avoided. Delivery failure occurs.
  • the device embodiment for information delivery described above can realize information delivery between multi-core systems or multi-core systems, high reliability of information delivery, improved speed and efficiency of information delivery, and load balancing of task instances in a multi-core system. It also prevents the failure of information delivery.
  • the letter configured in the multi-core system The delivery device independently maintains a preset global relationship table.
  • each multi-core system is configured with information delivery devices, and each information delivery device may separately maintain a preset global relationship table, or may select one of the information delivery devices. Maintain a preset global relationship table. If each information delivery device maintains a preset global relationship table, it is necessary to keep the preset global relationship table information in each information delivery device synchronized.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Hardware Redundancy (AREA)
  • Computer And Data Communications (AREA)
  • Debugging And Monitoring (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本发明公开了一种信息投递的方法和装置,其装置包括:接收模块,用于接收源任务发送的待投递信息和目标任务名称;查找模块,用于根据目标任务名称在预设全局关系表中查找与目标任务名称对应的内核号;发送模块,用于根据内核号在多核系统中查找对应的内核,并将待投递信息发送至所述内核内与目标任务名称对应的任务中。本发明提供的信息投递的方法和装置,可实现多核系统内或多核系统间信息的投递,信息投递可靠性高。

Description

信息投递的方法和装置 技术领域
本发明涉及多核技术领域, 尤其是涉及一种信息投递的方法和装置。 背景技术
在现有的多核系统中, 每个处理器是由多个内核组成, 每个内核包含 有一个或多个硬件线程。 在通信设备中由于业务的复杂性, 由多种业务模 块、 控制模块和交换单板组成, 每个单板釆用一个或多个多核处理器。
在现有的多核系统中的信息通信方式如图 1 所示, 该信息通信方式沿 用传统的单核系统处理模式,单板间和单板内的任务通过 IPC ( Inter-Process Communication, 进程间通信 )方式实现信息通信。 IPC方式基于 TCP/IP或 共享内存的方式实现信息通信, 然而在多核系统中, 这种信息通信方式无 法体现出多核系统的优势。 随着多核系统中内核数量的增加, 以内核为基 本信息通信单元的机制复杂度大大增加, 当大量信息汇聚到控制面单板需 处理时, 可能造成控制面单板处理能力过载, 使得信息投递的可靠性降低。 另夕卜, 该 IPC方式也无法实现多核系统处理的负载均衡, 这样也就无法实 现多核系统中任务多实例时的高效通信。 发明内容
有鉴于此, 本发明的主要目的在于提供一种的信息投递的方法和装置, 实现多核系统内或多核系统间信息的可靠投递。
为解决上述技术问题, 本发明提出一种信息投递的方法, 其包括步骤: 接收源任务发送的待投递信息和目标任务名称;
根据目标任务名称在预设全局关系表中查找与目标任务名称对应的内 核号;
根据内核号在多核系统中查找对应的内核, 并将待投递信息发送至所 述内核内与目标任务名称对应的任务中。
优选地, 所述预设全局关系表包含: 多核系统中的任务名称与该任务 所在内核的内核号对应关系。
优选地, 所述源任务与目标任务包括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核内、 或包括在不同多核系统的内核内。
优选地, 所述目标任务包括一个或多个目标任务实例; 多个目标任务 实例包括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内 核内、 或包括在不同多核系统的内核内。
优选地, 所述信息投递的方法, 当目标任务包括多个目标任务实例时, 则根据就近原则选择目标任务实例作为待投递信息的目标任务。
本发明另提出一种信息投递的装置, 其包括:
接收模块, 用于接收源任务发送的待投递信息和目标任务名称; 查找模块, 用于根据目标任务名称在预设全局关系表中查找与目标任 务名称对应的内核号;
发送模块, 用于根据内核号在多核系统中查找对应的内核, 并将待投 优选地, 所述预设全局关系表包含: 多核系统中的任务名称与该任务 所在内核的内核号对应关系。
优选地, 所述源任务与目标任务包括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核内、 或包括在不同多核系统的内核内。
优选地, 所述目标任务包括一个或多个目标任务实例; 多个目标任务 实例包括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内 核内、 或包括在不同多核系统的内核内。 所述查找模块, 还用于当目标任务包括多个目标任务实例时, 根据就 近原则选择目标任务实例作为待投递信息的目标任务。
由上可知, 利用本发明提供的信息投递的方法和装置, 可实现多核系 统内或多核系统间信息的投递, 信息投递可靠性高, 提高了信息投递的速 度和效率, 进一步地, 还实现了多核系统中任务实例的负载均衡, 同时还 可防止信息投递的失败。 附图说明
图 1是现有技术中多核系统结构示意图;
图 2是本发明的信息投递的方法一实施例流程示意图;
图 3 是本发明的多核系统结构示意图;
图 4 是本发明的多核系统间的结构示意图;
图 5是本发明的信息投递的装置结构示意图;
图 6是本发明的多核系统的另一结构示意图;
图 7是本发明的多核系统间的另一结构示意图。 具体实施方式
本发明目的的实现、 功能特点及优点将结合实施例, 参照附图做进一 步说明。 应当理解, 此处所描述的具体实施例仅仅用以解释本发明, 并不 用于限定本发明。
参见图 2, 提出本发明的信息投递的方法一实施例, 其包括: 步骤 101、 接收源任务发送的待投递信息和目标任务名称;
步骤 102、根据目标任务名称在预设全局关系表中查找与目标任务名称 对应的内核号;
步骤 103、才艮据内核号在多核系统中查找对应的内核, 并将待投递信息 发送至所述内核内与目标任务名称对应的任务中。 所述内核内与目标任务名称对应的任务即为目标任务。
进一步地, 上述信息投递的方法实施例中, 所述预设全局关系表包含: 多核系统中的任务名称与该任务所在内核的内核号对应关系。
上述信息投递的方法应用在多核系统中, 所述任务名称与该任务所在 内核的内核号对应关系, 是指多核系统内所有任务名称与该任务名称所在 内核的内核号对应关系, 或通讯连接的所有多核系统中的所有任务名称与 该任务所在内核的内核号对应关系。
进一步地, 上述信息投递的方法实施例中, 所述源任务与目标任务包 括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核内、 或包括在不同多核系统的内核内。
进一步地, 上述信息投递的方法实施例, 其中, 所述目标任务包括一 个或多个目标任务实例; 多个目标任务实例包括在同一多核系统的同一内 核内、 或包括在同一多核系统的不同内核内、 或包括在不同多核系统的内 核内。
参见图 3 , 设多核系统 1为独立的多核系统, 该多核系统 1内包括有: 内核①、 内核②和内核③。 其中, 内核①包括有任务 A-1和任务 C-1 ; 内核 ②包括有任务 A-2和任务 B-1 ; 内核③包括有任务 D-1和任务 E-l。 其预设 全局关系表包括有如下任务名称与该任务所在内核的内核号对应关系: A-l→①、 A-2→②、 B-l→②、 C-l→①、 D-l→③、 E-l→③。
参见图 4, 设多核系统 1、 2和 3为相互通讯的多核系统, 其中多核系 统 1包括有内核 i , 内核 i包括有任务 a-1和任务 b-1。 多核系统 2包括有 内核 ii和内核 iii , 内核 ii包括有任务 a-2和任务 d-1; 内核 iii包括有任务 a-3 和任务 c- 1。 多核系统 3包括有内核 iv和内核 V , 内核 iv包括有任务 b-2和 任务 d-2; 内核 V包括有任务 e-1 和任务 f—l。 其预设全局关系表包括有如 下任务名称与该任务所在内核的内核号对应关系: a-l→ i 、 a-2→ii 、 a-3→iii、 b-l→ i 、 b-2→iv、 c-l→iii、 d-l→ii、 d-2→iv、 e-l→ v 、 f-l→ v 。 进一步地, 上述信息投递的方法实施例中, 当目标任务只有一个目标 任务实例, 则直接将待投递信息发送至该目标任务实例中。
进一步地, 上述信息投递的方法实施例中, 当目标任务包括多个目标 任务实例时, 根据就近原则选择目标任务实例作为待投递信息的目标任务。 所述就近原则如下: 优先选择与源任务处于同一内核的目标任务实例作为 待投递信息的目标任务; 若没有与源任务处于同一内核的目标任务实例, 则优先选择与源任务处于同一多核系统的目标任务实例作为待投递信息的 目标任务; 若没有与源任务处于同一内核、 也没有与源任务处于同一多核 系统的目标任务实例, 则选择与源任务处于不同多核系统的目标任务实例 作为待投递信息的目标任务。
另外, 若源任务和目标任务在不同的多核系统中, 或在同一多核系统 中、 但位于不同内核中, 则可根据目标任务中的多个目标任务实例的负载 情况和是否异常的具体情况选择某一目标任务实例作为待投递信息的目标 任务。
以图 4为例, 设任务 d-1为源任务, 任务 d-1需向任务 a发送待投递信 息, 任务 a包括有 a-l、 a-2、 a-3三个任务实例。 其中, 任务 a-2处于内核 ii内, 任务 a-3处于内核 iii内, 任务 a-1处于内核 i内; 任务 d-1处于内核 ϋ内,根据上述的就近原则,优先选择与源任务 d-1处于同一内核的目标任 务实例 a-2作为待投递信息的目标任务。若被选为目标任务的目标任务实例 出现异常或负载较高的情况, 则按照同样原理在其他的目标任务实例中选 择一个作为待投递信息的目标任务。 即若目标任务实例 a-2 异常或负载较 高, 则优先选择与源任务 d-1处于同一多核系统的目标任务实例 a-3作为待 投递信息的目标任务; 若目标任务实例 a-2和 a-3都异常或负载较高, 则选 择与源任务处于不同多核系统的目标任务实例 a-i 作为待投递信息的目标 任务。
以图 4为例, 设任务 c-1为源任务, 任务 c-1需向任务 d发送待投递信 息。 其中, 任务 d包括有 d-1和 d-2两个任务实例; 源任务 c-1和任务 d-1 处于同一多核系统 2内;任务 d-2处于多核系统 3内;根据上述的就近原则, 优先选择任务 d-1作为待投递信息的目标任务;若任务 d-1异常或负载较高, 则选择任务 d-2作为待投递信息的目标任务。
以图 4为例, 设任务 d-2为源任务, 任务 d-2需向任务 a发送待投递信 息。 其中, 任务 a包括有 a-l、 a-2和 a-3三个任务实例。 任务 a的所有任务 实例 a-l、 a-2, a-3都不和源任务 d-2处于同一多核系统内; 任务 a-1处于 多核系统 1内, 而任务 a-2和 a-3处于多核系统 2内。 根据上述就近原则无 法选择, 此时需根据任务 a三个任务实例是否异常或负载高低的情况进行 选择, 将运行正常, 且负载较低的任务实例作为待投递信息的目标任务。
通过上述就近原则的选择方式, 可以提高信息投递的速度和效率, 可 避免信息投递失败现象发生, 同时还可保证同一任务中的不同任务实例的 负载均衡。
另外, 上述信息投递的方法实施例, 当目标任务包括多个目标任务实 例时, 还可不根据就近原则而直接查找负载最低的目标任务实例, 将该目 标任务实例作为待投递信息的目标任务。以图 4为例 ,设任务 a-1为源任务, 任务 a-1需向任务 b发送待投递信息, 任务 b包括有 b-1和 b-2两个任务实 例, 任务 b-1负载大, 任务 b-2负载低, 因此选择任务 b-2为待投递信息的 目标任务,将待投递信息发送至内核 iv内的任务 b-2中。若被选为目标任务 的目标任务实例出现异常情况, 则按照同样原理在其他的目标任务实例中 选择一个作为待投递信息的目标任务, 以保证同一任务中的不同任务实例 的负载均衡, 同时还可避免信息投递失败现象发生。
以上所述信息投递的方法实施例, 可实现多核系统内或多核系统间信 息的投递, 信息投递可靠性高, 提高了信息投递的速度和效率、 实现了多 核系统中任务实例的负载均衡, 同时还可防止信息投递的失败。
参见图 5, 提出本发明的信息投递的装置 10—实施例, 其包括: 接收 模块 11、 查找模块 12和发送模块 13。 其中, 接收模块 11 , 用于接收源任 务发送的待投递信息和目标任务名称; 查找模块 12, 用于根据目标任务名 称在预设全局关系表中查找与目标任务名称对应的内核号; 发送模块 13, 用于根据内核号在多核系统中查找对应的内核, 并将待投递信息发送至所 述内核内与目标任务名称对应的任务中。
所述内核内与目标任务名称对应的任务即为目标任务。
进一步地, 上述信息投递的装置实施例中, 所述预设全局关系表包含: 多核系统中的任务名称与该任务所在内核的内核号对应关系。
上述信息投递的装置 10应用在多核系统中, 所述任务名称与该任务所 在内核的内核号对应关系, 是指多核系统内所有任务名称与该任务名称所 在内核的内核号对应关系, 或通讯连接的所有多核系统中的所有任务名称 与该任务所在内核的内核号对应关系。
进一步地, 上述信息投递的装置实施例, 其中, 所述源任务与目标任 务包括在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核 内、 或包括在不同多核系统的内核内。
进一步地, 上述信息投递的装置实施例, 其中, 所述目标任务包括一 个或多个目标任务实例; 多个目标任务实例包括在同一多核系统的同一内 核内、 或包括在同一多核系统的不同内核内、 或包括在不同多核系统的内 核内。
参见图 6,设多核系统 1为独立的多核系统, 该多核系统配置有本发明 的信息投递的装置 10。 该多核系统 1内包括有: 内核①、 内核②和内核③。 其中, 内核①包括有任务 A-1和任务 C-1; 内核②包括有任务 A-2和任务 B-1 ; 内核③包括有任务 D-1和任务 E-l。 其预设全局关系表包括有如下任 务名称与该任务所在内核的内核号对应关系: A-l→①、 A-2→②、 B-l→②、 C-l→①、 D-l→③、 E-l→③。 源任务需要发送待投递信息时, 首先将待投 递信息和目标任务名称发送给信息投递的装置 10,信息投递的装置 10则根 据目标任务名称在预设全局关系表中确定与该目标任务名称对应的内核 号, 然后在多核系统 1 中查找与内核号对应的内核, 并将待投递信息发送 到该内核中与目标任务名称对应的任务中。
参见图 7 , 设多核系统 1、 2和 3为相互通讯的多核系统, 每个多核系 统中分别配置有本发明的信息投递的装置 10。 其中多核系统 1包括有内核 i , 内核 i包括有任务 a-1和任务 b-1。多核系统 2包括有内核 ii和内核 iii, 内核 ii包括有任务 a-2和任务 d-1; 内核 iii包括有任务 a-3和任务 c-1。 多核 系统 3包括有内核 iv和内核 V , 内核 iv包括有任务 b-2和任务 d-2; 内核 v 包括任务 e-1和任务 f-l。 其预设全局关系表包括有如下任务名称与该任务 所在内核的内核号对应关系: a-l→ i 、 a-2→ ii、 a-3→iii、 b-l→ i 、 b-2→iv、 c-l→iii、 d-l→ii、 d-2→iv、 e-l→v 、 f-l→v。 同理, 源任务需要发送待 投递信息时,首先将待投递信息和目标任务名称发送给信息投递的装置 10, 信息投递的装置 10则根据目标任务名称在预设全局关系表中确定与该目标 任务名称对应的内核号, 然后在多核系统 1、 2和 3中查找与内核号对应的 内核, 并将待投递信息发送到该内核中与目标任务名称对应的任务中。
进一步地, 上述信息投递的装置实施例中, 当目标任务只有一个目标 任务实例, 则直接将待投递信息发送至该目标任务实例中。
进一步地, 上述信息投递的装置实施例中, 所述查找模块 12, 还用于 当目标任务包括多个目标任务实例时, 根据就近原则选择目标任务实例作 为待投递信息的目标任务。 所述就近原则如下: 优先选择与源任务处于同 一内核的目标任务实例作为待投递信息的目标任务; 若没有与源任务处于 同一内核的目标任务实例, 则优先选择与源任务处于同一多核系统的目标 任务实例作为待投递信息的目标任务; 若没有与源任务处于同一内核、 也 没有与源任务处于同一多核系统的目标任务实例, 则选择与源任务处于不 同多核系统的目标任务实例作为待投递信息的目标任务。
另外, 若源任务和目标任务在不同的多核系统中, 或在同一多核系统 中、 但位于不同内核中, 信息投递的装置 10可根据目标任务中的多个目标 任务实例的负载情况和是否异常的具体情况选择某一目标任务实例作为待 投递信息的目标任务。
以图 7为例, 设任务 d-1为源任务, 任务 d-1需向任务 a发送待投递信 息, 任务 a包括有 a-l、 a-2、 a-3三个任务实例。 首先任务 d-1将待投递信 息和目标任务名称 a发送给信息投递的装置 10,信息投递的装置 10根据目 标任务名称 a在预设全局关系表中查找与该目标任务名称 a对应的内核号有 i 、 ii和 iii (包括目标任务 a的内核号)。 然后在多核系统 1、 2和 3中查 找与内核号 i 、 ii和 iii对应的内核 i 、 ii和 iii。 任务 a-2处于内核 ii内, 任 务 a-3处于内核 iii内, 任务 a-1处于内核 i内; 任务 d-1处于内核 ii内, 根 据上述的就近原则, 信息投递的装置 10优先选择与源任务 d-1处于同一内 核的目标任务实例 a-2作为待投递信息的目标任务,然后将待投递信息发送 至内核 ii内的任务 a-2中。 若被选为目标任务的目标任务实例 a-2出现异常 或负载较高的情况, 则按照同样原理在其他的目标任务实例中选择一个作 为待投递信息的目标任务。 即若目标任务实例 a-2异常或负载较高, 则信息 投递的装置 10优先选择与源任务 d-1处于同一多核系统的目标任务实例 a-3 作为待投递信息的目标任务, 然后将待投递信息发送至内核 iii内的任务 a-3 中; 若目标任务实例 a-2和 a-3都异常或负载较高, 则信息投递的装置 10 选择与源任务处于不同多核系统的目标任务实例 a-i 作为待投递信息的目 标任务, 然后将待投递信息发送至内核 i内的任务 a-1中。 即若被选为目标 任务的目标任务实例出现异常和负载较高情况, 且除被选目标任务实例还 有两个或两个以上目标任务实例, 则按照同样原理在剩余的目标任务实例 中选择一个作为待投递信息的目标任务, 以提高信息投递的速度和效率, 同时还可避免信息投递失败现象发生。
以图 7为例, 设任务 c-1为源任务, 任务 c-1需向任务 d发送待投递信 息, 任务 d包括有 d-1和 d-2两个任务实例。 首先任务 c-1将待投递信息和 目标任务名称 d发送给信息投递的装置 10,信息投递的装置 10根据目标任 务名称 d在预设全局关系表中查找与该目标任务名称 d对应的内核号有 ii 和 iv (包括目标任务 d的内核号)。 然后在多核系统 1、 2和 3中查找与内 核号 ii和 iv对应的内核 ii和 iv。 源任务 c-1处于内核 iii, 任务 d-1处于内核 ii , 内核 ii和内核 iii处于同一多核系统 2内;任务 d-2处于多核系统 3内的 内核 iv中; 根据上述的就近原则, 信息投递的装置 10优先选择任务 d-1作 为待投递信息的目标任务, 然后将待投递信息发送至任务 d-1 ; 若任务 d-1 异常或负载较高, 信息投递的装置 10则选择任务 d-2作为待投递信息的目 标任务, 然后将待投递信息发送至任务 d-2。
以图 7为例, 设任务 d-2为源任务, 任务 d-2需向任务 a发送待投递信 息。 其中, 任务 a包括有 a-l、 a-2和 a-3三个任务实例。 首先任务 d-2将待 投递信息和目标任务名称 a发送给信息投递的装置 10, 信息投递的装置 10 根据目标任务名称 a在预设全局关系表中查找与该目标任务名称 a对应的内 核号有 i 、 ii和 iii (包括目标任务 a的内核号)。 然后在多核系统 1、 2和 3 中查找与内核号 i 、 ii和 iii对应的内核 i 、 ii和 iii。任务 a的所有实例 a-l、 a-2、 a-3都不和源任务 d-2处于同一内核内, 也不处于同一多核系统内; 任 务 a-1处于多核系统 1内, 而任务 a-2和 a-3处于多核系统 2内。 信息投递 的装置 10无法根据上述就近原则进行选择, 此时需根据任务 a三个任务实 例是否异常或负载高低的情况进行选择, 将运行正常, 且负载较低的任务 实例作为待投递信息的目标任务。
信息投递的装置 10通过上述就近原则进行目标任务的选择, 可以提高 信息投递的速度和效率, 可避免信息投递失败现象发生, 同时还可保证同 一任务中的不同任务实例的负载均衡。
另外, 上述信息投递的装置 10实施例, 当目标任务包括多个目标任务 实施例时, 还可不根据就近原则而直接查找负载最低的目标任务实例, 将 该目标任务实例作为待投递信息的目标任务。 以图 7为例,设任务 a-1为源 任务, 任务 a-1需向任务 d发送待投递信息, 首先任务 a-1将待投递信息和 目标任务名称 d发送给信息投递的装置 10,信息投递的装置 10根据目标任 务名称 d在预设全局关系表中查找与该目标任务名称 d对应的内核号有 ii 和 iv (包括目标任务 d的内核号)。 然后在多核系统 1、 2和 3中查找与内 核号 ii、 iv对应的内核 ii和 iv , 任务 d的 d-1和 d-2两个任务实例, 其中任 务 d-1处于内核 ii内, 任务 d-2处于内核 iv内, 任务 d-1负载大, 任务 d-2 负载低, 因此选择内核 iv的任务 d-2为待投递信息的目标任务,信息投递的 装置 10则将待投递信息发送至内核 iv内的任务 d-2中。 若被选为目标任务 的目标任务实例 d-2出现异常情况,则信息投递的装置将待投递信息发送至 内核 ii的任务 d-1中。若被选为目标任务的目标任务实例出现异常情况,且 在其他的目标任务实例中选择一个作为待投递信息的目标任务, 以保证同 一任务中的不同任务实例的负载均衡, 同时还可避免信息投递失败现象发 生。
以上所述信息投递的装置实施例, 可实现多核系统内或多核系统间信 息的投递, 信息投递可靠性高, 提高了信息投递的速度和效率、 实现了多 核系统中任务实例的负载均衡, 同时还可防止信息投递的失败。
本发明中, 当多核系统为单独的多核系统, 则配置在该多核系统的信 息投递的装置独立维护一张预设全局关系表。 当多核系统为多个互为通讯 连接多核系统, 每个多核系统分别配置信息投递的装置, 每个信息投递的 装置可以分别维护一张预设全局关系表, 也可以选择其中一个信息投递的 装置维护一张预设全局关系表。 若每个信息投递的装置分别维护一张预设 全局关系表, 则需保持每个信息投递的装置中预设全局关系表信息同步。
应当理解的是, 以上仅为本发明的优选实施例, 不能因此限制本发明 的专利范围, 凡是利用本发明说明书及附图内容所作的等效结构或等效流 程变换, 或直接或间接运用在其他相关的技术领域, 均同理包括在本发明 的专利保护范围内。

Claims

权利要求书
1、 一种信息投递的方法, 其特征在于, 包括步骤:
接收源任务发送的待投递信息和目标任务名称;
根据目标任务名称在预设全局关系表中查找与目标任务名称对应的内 核号;
根据内核号在多核系统中查找对应的内核, 并将待投递信息发送至所 述内核内与目标任务名称对应的任务中。
2、 根据权利要求 1所述的信息投递的方法, 其特征在于,
所述预设全局关系表包含: 多核系统中的任务名称与该任务所在内核 的内核号对应关系。
3、 根据权利要求 2所述的信息投递的方法, 其特征在于,
所述源任务与目标任务包括在同一多核系统的同一内核内、 或包括在 同一多核系统的不同内核内、 或包括在不同多核系统的内核内。
4、 根据权利要求 1至 3任一项所述的信息投递的方法, 其特征在于, 所述目标任务包括一个或多个目标任务实例; 多个目标任务实例包括 在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核内、 或 包括在不同多核系统的内核内。
5、 根据权利要求 4所述的信息投递的方法, 其特征在于,
当目标任务包括多个目标任务实例时, 则根据就近原则选择目标任务 实例作为待投递信息的目标任务。
6、 一种信息投递的装置, 其特征在于, 包括:
接收模块, 用于接收源任务发送的待投递信息和目标任务名称; 查找模块, 用于根据目标任务名称在预设全局关系表中查找与目标任 务名称对应的内核号;
发送模块, 用于根据内核号在多核系统中查找对应的内核, 并将待投
7、 根据权利要求 6所述的信息投递的装置, 其特征在于, 所述预设全局关系表包含: 多核系统中的任务名称与该任务所在内核 的内核号对应关系。
8、 根据权利要 7所述的信息投递的装置, 其特征在于,
所述源任务与目标任务包括在同一多核系统的同一内核内、 或包括在 同一多核系统的不同内核内、 或包括在不同多核系统的内核内。
9、 根据权利要求 6至 8任一项所述的信息投递的装置, 其特征在于, 所述目标任务包括一个或多个目标任务实例; 多个目标任务实例包括 在同一多核系统的同一内核内、 或包括在同一多核系统的不同内核内、 或 包括在不同多核系统的内核内。
10、 根据权利要求 9所述的信息投递的装置, 其特征在于,
所述查找模块, 还用于当目标任务包括多个目标任务实例时, 根据就 近原则选择目标任务实例作为待投递信息的目标任务。
PCT/CN2011/075893 2010-10-25 2011-06-17 信息投递的方法和装置 WO2012055248A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP11835522.1A EP2634694B1 (en) 2010-10-25 2011-06-17 Method and device for information delivery
US13/881,018 US9207974B2 (en) 2010-10-25 2011-06-17 Information delivery method and device
RU2013121783/08A RU2566953C2 (ru) 2010-10-25 2011-06-17 Способ и устройство для доставки информации
BR112013010079-6A BR112013010079A2 (pt) 2010-10-25 2011-06-17 método e dispositivo de fornecimento de informação

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010518285.XA CN101976208B (zh) 2010-10-25 2010-10-25 信息投递的方法和装置
CN201010518285.X 2010-10-25

Publications (1)

Publication Number Publication Date
WO2012055248A1 true WO2012055248A1 (zh) 2012-05-03

Family

ID=43576095

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/075893 WO2012055248A1 (zh) 2010-10-25 2011-06-17 信息投递的方法和装置

Country Status (7)

Country Link
US (1) US9207974B2 (zh)
EP (1) EP2634694B1 (zh)
CN (1) CN101976208B (zh)
BR (1) BR112013010079A2 (zh)
RU (1) RU2566953C2 (zh)
TR (1) TR201806949T4 (zh)
WO (1) WO2012055248A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976208B (zh) 2010-10-25 2014-08-13 中兴通讯股份有限公司 信息投递的方法和装置
CN102681905A (zh) * 2011-03-09 2012-09-19 新奥特(北京)视频技术有限公司 一种任务之间的相互通讯方法及系统
CN104572315B (zh) * 2013-10-23 2018-12-25 华为技术有限公司 实现子系统间通信的方法、通信实体及分布式通信系统
CN107786544B (zh) * 2017-09-29 2018-08-17 贵州白山云科技有限公司 一种消息的任务状态处理方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101110043A (zh) * 2007-09-04 2008-01-23 杭州华三通信技术有限公司 一种多核系统的资源管理方法和控制核
US20090094437A1 (en) * 2007-10-07 2009-04-09 Masahiro Fukuda Method And Device For Controlling Multicore Processor
CN101634953A (zh) * 2008-07-22 2010-01-27 国际商业机器公司 搜索空间计算方法和装置及自适应线程调度方法和系统
CN101976208A (zh) * 2010-10-25 2011-02-16 中兴通讯股份有限公司 信息投递的方法和装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7234139B1 (en) 2000-11-24 2007-06-19 Catharon Productions, Inc. Computer multi-tasking via virtual threading using an interpreter
JP2004334537A (ja) * 2003-05-07 2004-11-25 Sony Corp プログラム処理システム及びプログラム処理方法、並びにコンピュータ・プログラム
EP1987428A2 (en) * 2006-02-24 2008-11-05 Mbda Uk Limited Real-time distributed processor environment
CN100466629C (zh) * 2006-09-18 2009-03-04 杭州华三通信技术有限公司 一种网络设备和基于多核处理器的报文转发方法
CN101262425A (zh) * 2008-04-28 2008-09-10 艾诺通信系统(苏州)有限责任公司 基于网络地址转换的多播转发的方法
CN101403982B (zh) * 2008-11-03 2011-07-20 华为技术有限公司 一种多核处理器的任务分配方法和系统
US20100287320A1 (en) * 2009-05-06 2010-11-11 Lsi Corporation Interprocessor Communication Architecture
US8601556B2 (en) * 2009-06-22 2013-12-03 Citrix Systems, Inc. Systems and methods for handling SSL session not reusable across multiple cores

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101110043A (zh) * 2007-09-04 2008-01-23 杭州华三通信技术有限公司 一种多核系统的资源管理方法和控制核
US20090094437A1 (en) * 2007-10-07 2009-04-09 Masahiro Fukuda Method And Device For Controlling Multicore Processor
CN101634953A (zh) * 2008-07-22 2010-01-27 国际商业机器公司 搜索空间计算方法和装置及自适应线程调度方法和系统
CN101976208A (zh) * 2010-10-25 2011-02-16 中兴通讯股份有限公司 信息投递的方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2634694A4 *

Also Published As

Publication number Publication date
RU2566953C2 (ru) 2015-10-27
BR112013010079A2 (pt) 2020-07-07
CN101976208B (zh) 2014-08-13
US20130227580A1 (en) 2013-08-29
TR201806949T4 (tr) 2018-06-21
RU2013121783A (ru) 2014-12-10
EP2634694A4 (en) 2016-05-25
EP2634694B1 (en) 2018-04-04
US9207974B2 (en) 2015-12-08
EP2634694A1 (en) 2013-09-04
CN101976208A (zh) 2011-02-16

Similar Documents

Publication Publication Date Title
CN104620558B (zh) 用于支持分布式数据网格集群中的消息预处理的系统和方法
CN112313916B (zh) 一种融合区块链技术拟态存储防篡改日志的方法及系统
CN1906586B (zh) 用于在多处理器系统中处置处理错误的方法和设备
CN108200124B (zh) 一种高可用应用程序架构及构建方法
US10826977B2 (en) System and method for supporting asynchronous request/response in a network environment
WO2012055248A1 (zh) 信息投递的方法和装置
CN105610972A (zh) 集群式的任务调派系统
US9411633B2 (en) System and method for barrier command monitoring in computing systems
CN105183549A (zh) 基于任务分配的自动出票系统
CN102567227A (zh) 共享缓存设备的双控制器存储系统和方法
CN102025783A (zh) 集群系统及其消息处理方法和协议转发网关
CN105790825A (zh) 一种分布式保护中控制器热备份的方法和装置
EP2761453B1 (en) System and method for preventing single-point bottleneck in a transactional middleware machine environment
CN101686261A (zh) 一种基于rac的冗余服务器系统
CN102231126B (zh) 一种实现多核处理器中核间备份的方法及系统
CN103179162B (zh) 一种输出日志的方法及系统
CN114371935A (zh) 网关处理方法、网关、设备及介质
CN110046064B (zh) 一种基于故障漂移的云服务器容灾实现方法
CN106294031A (zh) 一种业务管理方法和存储控制器
TW201140302A (en) Data storage system with power backup mechanism
US9785594B2 (en) Micro server and switch device thereof
JP5061674B2 (ja) フォールトトレラントコンピュータシステム及びデータ送信制御方法
CN104765420A (zh) 一种基于Power平台的内存板方案设计方法
CN102662911A (zh) 一种板级重构红外信号处理机的控制方法
Yu et al. Design and implementation of open mpi over quadrics/elan4

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11835522

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13881018

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2011835522

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2013121783

Country of ref document: RU

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112013010079

Country of ref document: BR

REG Reference to national code

Ref country code: BR

Ref legal event code: B01E

Ref document number: 112013010079

Country of ref document: BR

Free format text: APRESENTE A TRADUCAO SIMPLES DA FOLHA DE ROSTO DA CERTIDAO DE DEPOSITO DA PRIORIDADE CN 201010518285.X; OU DECLARACAO DE QUE OS DADOS DO PEDIDO INTERNACIONAL ESTAO FIELMENTE CONTIDOS NA PRIORIDADE REIVINDICADA, CONTENDO TODOS OS DADOS IDENTIFICADORES DESTA (TITULARES,NUMERO DE REGISTRO, DATA E TITULO), CONFORME O PARAGRAFO UNICO DO ART. 25 DA RESOLUCAO T7/2013 E ART. 2O DA RESOLUCAO 179/2017. CABE SALIENTAR QUE NAO FOI POSSIVEL IDENTIFICAR O TITULAR DO PEDIDO PRIORITARIO NOS DOCUMENTOS JUNTADOS AO PROCESSO, TAMPOUCO NOS APRESENTADOS NA OMPI. TAL INFORMACAO E NECESSARIA PARA O EXAME DA CESSAO DO DOCUMENTO DE PRIORIDADE.

ENP Entry into the national phase

Ref document number: 112013010079

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20130425