WO2011053038A2 - 교착 상태의 방지를 위한 데이터 처리 방법 및 시스템 - Google Patents
교착 상태의 방지를 위한 데이터 처리 방법 및 시스템 Download PDFInfo
- Publication number
- WO2011053038A2 WO2011053038A2 PCT/KR2010/007517 KR2010007517W WO2011053038A2 WO 2011053038 A2 WO2011053038 A2 WO 2011053038A2 KR 2010007517 W KR2010007517 W KR 2010007517W WO 2011053038 A2 WO2011053038 A2 WO 2011053038A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- message
- actor
- read
- received
- read message
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000004044 response Effects 0.000 claims abstract description 65
- 230000008569 process Effects 0.000 claims abstract description 19
- 238000003672 processing method Methods 0.000 claims abstract description 14
- 238000004148 unit process Methods 0.000 claims 1
- 230000001360 synchronised effect Effects 0.000 description 16
- 230000005540 biological transmission Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 230000009471 action Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
- G06F9/524—Deadlock detection or avoidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
Definitions
- the present invention relates to a data processing method and system for preventing a deadlock, and more particularly, a data processing method capable of preventing a deadlock by processing a message received in a message response waiting state in an actor system. And to the system.
- An actor system as a method for concurrent computation in a distributed system or a multithreading system.
- An actor system is a system that sees an actor as the basic element of parallel computation.
- An actor can receive and process a message, make calculations and make decisions on its own, create another actor, or send a message to another actor.
- each actor has its own state, that is, its own performer for processing data and data, for example, a thread, and interacts asynchronously by exchanging messages with other actors.
- An actor can perform actions such as 1) modify its own state (data), 2) send a message to another actor, 3) create a new actor, or 4) transfer to another host.
- Synchronization messages can be used in actor systems.
- a synchronous message is a message that checks the response to a sent message and then performs the next action accordingly.
- Synchronous message method is widely used in server / client system including distributed system. Direct-call in general application system can be seen as synchronous message method.
- Allowing these synchronous messages generally without restriction can result in deadlocks, where two actors wait for a long time asking each other for a response to each other's lead messages, preventing them from performing other processing.
- the occurrence of such a deadlock can cause serious problems such as the entire system freezing.
- FIG. 1 is a signal flow diagram illustrating a flow of a data processing method through transmission and reception of a synchronization message in an actor system according to the prior art.
- step S10 the first actor 10 calculates a1, which is an element for data processing, and requests b1, which is another element required for data processing.
- step S15 the first actor 10 searches for a second actor that is an actor to request b1 and generates a first read message for requesting b1 to the second actor.
- the first read message generated by the first actor 10 is a synchronization message.
- the first read message is a message of a method of checking a response to the transmitted message and then performing the following procedure.
- step S20 the first actor 10 sends a first read message to the second actor 20.
- step S25 the first actor 10 sets a response wait state corresponding to the transmission of the first read message to the second actor 20. That is, since the first read message is a synchronization message, the first actor 10 may check the response of the second actor 20 to the first read message and then perform the following procedure.
- the first actor 10 sets the state of the first actor 10 to a response waiting state until the response of the first read message sent to the second actor 20 is received.
- step S30 the second actor 20 stores the first read message received from the first actor 10 in a message queue.
- step S35 the second actor 20 transmits a second read message to the first actor 10 to obtain state information of the first actor 10.
- the second read message is a message not related to the first read message.
- step S40 the second actor 20 sets a response wait state corresponding to the transmission of the second read message to the first actor 10.
- the second actor 20 can perform the following procedure only after confirming the response of the first actor 10 to the second read message.
- the first actor 10 and the second actor 20 are set in a waiting state until receiving a response from each other, so that they cannot process other messages.
- the first actor 10 may not transmit the response of the second lead message to the second actor 20 and the second actor 20 transmits the response of the first lead message to the first actor 10. You won't be able to.
- An embodiment of the present invention prevents deadlocks caused by waiting for a long time for a response of a synchronous message by processing a synchronous message even before sending a synchronous message and receiving a synchronous message between actors. I would like to.
- the actor when an actor exists in the same memory space, the actor reads state information of a target actor in a direct call manner to improve data processing speed and performance.
- the first aspect of the present invention (a) transmitting a read message for requesting the status information of the target actor required for processing the data, (b) the lead Receiving and processing only a read message that reads the status information of the actor until a response of the message is received; and (c) if the response of the read message is received, processing related data using the status information. It may provide a data processing method for preventing a deadlock comprising a.
- the read message processed in step (b) is received from a message queue, and the message queue may store an event message or a read message for changing an actor's state.
- a second aspect of the present invention provides a data processor for processing data using state information of a target actor, a read message generator for generating a read message requesting the state information to the target actor, and the target. It is possible to provide a data processing system for preventing a deadlock including a reception message processing unit which processes only a read message that reads the status information of the actor among the received messages until a response to the read message sent to the actor is received. have.
- the read message received from another actor can be processed until the read message, which is a sync message, and the response to the read message is processed, so that the read message can be processed.
- the deadlock can be prevented by dispatching and processing from a message queue or the like without waiting for a read message received while waiting.
- FIG. 1 is a signal flow diagram illustrating a flow of a data processing method through transmission and reception of a synchronization message in an actor system according to the prior art
- FIG. 2 is a signal flow diagram illustrating a flow of a data processing method according to an embodiment of the present invention
- FIG. 3 is a signal flow diagram illustrating a flow of a data processing method according to an embodiment of the present invention.
- FIG. 4 is a flowchart illustrating a message processing flow of an actor in a data processing method according to an embodiment of the present invention
- FIG. 5 is a diagram showing the configuration of a data processing system according to an embodiment of the present invention.
- FIG. 6 is a diagram showing the configuration of an actor in a data processing system according to an embodiment of the present invention.
- FIGS. 2 and 3 are signal flow diagrams illustrating the flow of a data processing method according to an embodiment of the present invention.
- step S100 the first actor 100 calculates a1, which is an element for data processing, and requests b1, which is another element required for data processing.
- b1 is an element that should be received from actors other than the first actor 100 and may include state information of the actor.
- step S105 the first actor 100 generates a first read message for requesting b1 from the second actor 200.
- the first read message generated by the first actor 100 is a synchronous message.
- the first read message is a message in which a response to the transmitted message is checked and the following procedure is performed.
- Synchronous message method is widely used in server / client system including distributed system, and direct-call in general application system can be seen as synchronous message method.
- step S110 the first actor 100 transmits a first read message to the second actor 200.
- the first read message may be transmitted from the first actor 100 to the second actor 200 by the message performer.
- the message performer obtains the message from the message queue, delivers the message to the appropriate actor, and executes the content of the message.
- a message performer may be a thread provided by an operating system (OS), and a message performer may exist for each actor.
- OS operating system
- Threads may be executed in different central processing unit (CPU) cores in one host and may be distributed to different hosts.
- CPU central processing unit
- each actor includes one independent message performer for each actor.
- step S115 the second actor 200 stores the first read message received from the first actor 100 in a message queue.
- the message queue is a queue of messages waiting to be performed.
- the message queue may have a single message queue on one host, and may include a message queue for each actor. In the description of one embodiment according to the present invention, it is assumed that each actor includes a message queue for convenience of understanding.
- step S120 the second actor 200 transmits a second read message to the first actor 100.
- step S125 the first actor 100 processes the second read message received from the second actor 200.
- the first actor 100 is not set to a response wait state even when a read message is transmitted to the second actor 200.
- the first actor 100 may process the second read message even if the first actor message does not receive a response from the second actor 200.
- step S130 the first actor 100 transmits a second read message response including the processing result in step S125 to the second actor 200.
- step S135 the second actor 200 or the third actor 200 transmits a third read message to the first actor 100, and in step S140 the first actor 100
- the second read message received from the second actor 200 or the third actor 200 is processed, and in step S145, the first actor 100 includes the third read including the processing result in step S140.
- the read message response is transmitted to the second actor 200 or the third actor 200.
- the first actor 100 since the first actor 100 does not set the response waiting state even when the first lead message is transmitted to the second actor 200, the first actor 100 processes and responds to the read message received from one or more other actors. You can send
- step S150 the second actor 200 analyzes the first read message to calculate element b1, which is state information of the second actor 200 requested by the first actor 100.
- the second actor 200 transmits a response of the first read message to the first actor 100.
- the response of the first read message transmitted by the second actor 200 includes the state information b1 of the second actor 200 calculated in step S150.
- step S160 the first actor 100 processes data using element a1 calculated in step S100 and element b1 included in the response of the first read message received from the second actor 200. .
- FIG. 4 is a flowchart illustrating a message processing flow of an actor in a data processing method according to an embodiment of the present invention.
- step S200 it is determined that a situation occurs that requires status information of another actor, that is, a target actor. In other words, while performing data processing, it is determined that the status information of another actor is needed for data processing.
- step S210 it is determined whether the target actor that is the target of the requested state information exists in the same memory space.
- step S220 when it is determined in step S210 that the target actor does not exist in the same memory space, a read message for reading the status information of the target actor may be generated and processed as a target actor for data processing. Send the created lead message.
- the messages generated and transmitted by the actor include event messages or event messages.
- An event message is an asynchronous message in which the state of the receiving actor can be changed
- a read message is a synchronous message that only reads the state of the receiving actor. Therefore, in order to read the status information of the target actor, a read message is generated and transmitted to the target actor.
- Transmission of the read message may be performed by the message performer. Threads provided by the operating system (OS) may act as message performers.
- OS operating system
- step S230 it is determined whether a response to the read message transmitted in step S220 is received.
- the read message is a synchronous message that checks the response and then performs the corresponding procedure. Accordingly, by determining whether a response of the transmitted read message is received, it may be determined whether to perform a subsequent procedure according to receiving the response.
- step S240 when it is determined in step S230 that no response is received, the read message stored in the message queue is dispatched from the message queue.
- the message queue stores the received message and waits for it to be processed by the actor.
- the actor may process the received message immediately, but in some situations it may be necessary to wait for the received message to perform other processing.
- the actor may store the received message in the message queue, wait for the message, and dispatch the message from the message queue when the processing time is available.
- step S250 the read message dispatched in step S240 is analyzed and processed.
- the lead message is analyzed and processed according to the analyzed lead message, and the response is transmitted to the corresponding actor.
- an actor may process a read message received from another actor during a response waiting time until the read message is transmitted and the response is received.
- the actor's state when sending the lead message and the actor's state when receiving the response are changed. This can be different and can cause problems when processing data in response to the read message sent.
- the message queue may include an event message which is an asynchronous message as well as a read message which is a synchronous message, but a message that can be processed while waiting for a response of a transmitted read message is a synchronous that does not change the state of an actor It is limited to a read message which is a message.
- step S260 it is determined whether a response to the read message transmitted in step S220 is received.
- step S270 when the response of the read message sent in step S220 is received, the data is processed using the state information of the target actor included in the received response.
- step S280 when it is determined in step S210 that the target actor exists in the same memory space, the state information of the target actor is read through a direct call method.
- step S290 the relevant data is processed using the state information read in step S280.
- This method of directly reading the status information of the target actor can improve the data processing speed and improve the performance of the system compared to the method of using the synchronous message that needs to wait for a response after sending the message.
- FIG. 5 is a diagram illustrating a configuration of a data processing system according to an embodiment of the present invention.
- each actor 100 and 200 includes a message performer 101 and 201 and a message queue 102 and 202, respectively.
- the first actor 100 determines that the state information of the second actor 200 is necessary for data processing, the first actor 100 generates a first read message for reading the state information of the second actor 200, The generated first read message is transmitted to the second actor 200.
- the first actor 100 receives a read message and an event message from the second actor 200 or other actors, and if it cannot process the received message immediately, the first actor 100 sends a message performer 101. Store the received message in the message queue 102 via the.
- the first actor 100 dispatches the messages stored in the message queue 102 sequentially using the message performer 101 and performs a procedure according to the type and content of the messages.
- the first actor 100 when dispatching a read message, the first actor 100 generates a response including the state information of the first actor required by the read message, and generates the response as the actor that transmitted the read message. Send a response.
- the first actor 100 When dispatching the event message, the first actor 100 changes the state of the first actor 100 according to the information included in the event message.
- the first actor 100 receives the response of the first read message from the second actor 200 to receive the state information of the second actor 200, and performs data processing using the received state information.
- an actor in the data processing system can process the read message received even when the response of the read message sent to the target actor is not received. Deadlocks that occur can be prevented.
- FIG. 6 is a diagram illustrating a configuration of an actor of a data processing system according to an exemplary embodiment of the present invention.
- the actor 100 includes a signal transceiver 110, a data processor 120, a read message generator 130, and a received message processor 140.
- the signal transmitting and receiving unit 110 transmits a signal such as a message to another actor and receives the signal.
- the signal transceiver 110 receives a message such as a read message or an event message and transmits a signal including a response of the received read message to the corresponding actor. In addition, the signal transceiver 110 transmits a read message to the target actor and receives a response of the read message.
- the data processor 120 determines whether the actor exists in the same memory space as the actor 100 when the state information of the target actor is required for data processing. If the actor does not exist in the same memory space, the data processor 120 requests the read message generator 130 to generate a read message to be transmitted to the actor.
- the data processing unit 120 When the data processing unit 120 receives the transmission status from the lead message generating unit 130 indicating that the lead message has been transmitted to the corresponding actor, the data processing unit 120 processes only the read message and stops processing the event message until a response is received from the corresponding actor. Request to the received message processor 140.
- the data processor 120 requests the message processor 140 to process only the read message that does not change the state of the actor.
- the data processor 120 determines that the actor exists in the same memory space as the actor 100, the data processor 120 reads the state information of the actor in a direct call manner.
- the data processor 120 receives a response of the transmitted read message from the actor, and performs data processing using the state information included in the received response.
- the read message generator 130 generates a read message to be transmitted in response to a request of the data processor 120.
- the lead message generator 130 transmits a lead message generated using the information of the actor to receive the lead message included in the request of the data processor 120 through the signal transceiver 110, and transmits the transmission status to the data processor. Notify at 120.
- the received message processor 140 processes a message by dispatching a message stored in the message queue 102.
- the reception message processing unit 140 stops processing the event message and dispatches only the read message to process it.
- reception message processing unit 140 processes the message only to the read message until the actor 100 receives the response of the read message sent to the target actor.
- the reception message processor 140 may process the received read message even before receiving a response of the read message sent to the target actor, thereby preventing the message queue from waiting excessively.
- the above-described embodiment of the present invention can be applied not only to an environment distributed to a plurality of hosts, but also to one host, especially one or more CPU cores in one host, so-called multi-core environment.
- the present invention can be applied to the situation where the actors exist on the same host as well as the situation where each actor exists on a separate host.
- One embodiment of the present invention can also be implemented in the form of a recording medium containing instructions executable by a computer, such as a program module executed by the computer.
- Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media.
- Computer readable media may include both computer storage media and communication media.
- Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.
- Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transmission mechanism, and includes any information delivery media.
- each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
- Multi Processors (AREA)
- Computer And Data Communications (AREA)
Abstract
Description
Claims (11)
- 교착 상태의 방지를 위한 데이터 처리 방법에 있어서,(a) 데이터의 처리에 필요한 대상 액터의 상태 정보를 요청하는 리드(read) 메시지를 전송하는 단계,(b) 상기 리드 메시지의 응답을 수신할 때까지, 액터의 상태 정보를 판독하는 리드 메시지만을 수신하여 처리하는 단계 및(c) 상기 리드 메시지의 응답을 수신한 경우, 상기 상태 정보를 이용하여 관련 데이터를 처리하는 단계를 포함하는 데이터 처리 방법.
- 제 1 항에 있어서,(d) 상태 정보가 필요한 대상 액터가 동일한 메모리 공간에 존재하는지 여부를 판단하는 단계,(e) 상기 대상 액터가 동일한 메모리 공간에 존재하는 경우, 상기 대상 액터로부터 상기 상태 정보를 직접 판독하는 단계 및(f) 상기 직접 판독한 상태 정보를 이용하여 관련 데이터를 처리하는 단계를 더 포함하는 데이터 처리 방법.
- 제 1 항에 있어서,상기 (b) 단계에서 처리하는 리드 메시지는 메시지 큐(queue)로부터 수신되고,상기 메시지 큐는 액터의 상태를 변경시키는 이벤트 메시지 또는 리드 메시지를 저장하는 것인 데이터 처리 방법.
- 제 1 항에 있어서,상기 (b) 단계에서 처리하는 리드 메시지는 액터로부터 직접 수신되는 것인 데이터 처리 방법.
- 제 1 항에 있어서,리드 메시지의 전송 및 수신은 오퍼레이팅 시스템(OS, operating system)의 스레드(thread)를 통해 수행되는 것인 데이터 처리 방법.
- 제 1 항에 있어서,상기 대상 액터는 별도의 호스트(host)에 존재하는 것인 데이터 처리 방법.
- 제 1 항에 있어서,상기 대상 액터는 동일한 호스트(host)에 존재하는 것인 데이터 처리 방법.
- 교착 상태의 방지를 위한 데이터 처리 시스템에 있어서,대상 액터의 상태 정보를 이용하여 데이터를 처리하는 데이터 처리부,상기 상태 정보를 상기 대상 액터로 요청하는 리드(read) 메시지를 생성하는 리드 메시지 생성부 및상기 대상 액터로 전송된 상기 리드 메시지에 대한 응답이 수신될 때까지 수신한 메시지 중에서 액터의 상태 정보를 판독하는 리드 메시지만을 처리하는 수신 메시지 처리부를 포함하는 데이터 처리 시스템.
- 제 8 항에 있어서,상기 데이터 처리부는 동일한 메모리 공간에 존재하는 대상 액터의 상태 정보를 직접 수행(direct call) 방식으로 판독하여 데이터를 처리하는 것인 데이터 처리 시스템.
- 제 8 항에 있어서,상기 수신 메시지 처리부가 수신하는 리드 메시지는 액터로부터 직접 수신되는 것인 데이터 처리 시스템.
- 제 8 항에 있어서,상기 수신 메시지 처리부가 수신하는 리드 메시지는 액터의 상태를 변경시키는 이벤트(event) 메시지 또는 리드(read) 메시지를 저장하는 메시지 큐(queue)로부터 수신되는 것인 데이터 처리 시스템.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201080049494.2A CN102630315B (zh) | 2009-10-30 | 2010-10-29 | 用于防止死锁状态的数据处理方法及系统 |
US13/505,170 US8826301B2 (en) | 2009-10-30 | 2010-10-29 | Method and system for processing data for preventing deadlock |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2009-0104500 | 2009-10-30 | ||
KR1020090104500A KR101080845B1 (ko) | 2009-10-30 | 2009-10-30 | 교착 상태의 방지를 위한 데이터 처리 방법 및 시스템 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2011053038A2 true WO2011053038A2 (ko) | 2011-05-05 |
WO2011053038A3 WO2011053038A3 (ko) | 2011-11-03 |
Family
ID=43922863
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2010/007517 WO2011053038A2 (ko) | 2009-10-30 | 2010-10-29 | 교착 상태의 방지를 위한 데이터 처리 방법 및 시스템 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8826301B2 (ko) |
KR (1) | KR101080845B1 (ko) |
CN (1) | CN102630315B (ko) |
WO (1) | WO2011053038A2 (ko) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021000656A1 (zh) * | 2019-07-04 | 2021-01-07 | 中兴通讯股份有限公司 | 死锁检测控制方法、装置、通信设备及计算机存储介质 |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101191727B1 (ko) | 2011-02-28 | 2012-11-21 | (주)지노게임즈 | 멀티 버추얼 머신을 이용한 멀티 쓰레드 처리시스템 및 그 방법 |
CN103399818B (zh) * | 2013-08-13 | 2016-05-18 | 中国科学技术大学苏州研究院 | 操作系统中的死锁检测方法 |
RU2718215C2 (ru) | 2018-09-14 | 2020-03-31 | Общество С Ограниченной Ответственностью "Яндекс" | Система обработки данных и способ обнаружения затора в системе обработки данных |
RU2731321C2 (ru) | 2018-09-14 | 2020-09-01 | Общество С Ограниченной Ответственностью "Яндекс" | Способ определения потенциальной неисправности запоминающего устройства |
RU2714219C1 (ru) | 2018-09-14 | 2020-02-13 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для планирования передачи операций ввода/вывода |
RU2714602C1 (ru) | 2018-10-09 | 2020-02-18 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для обработки данных |
RU2721235C2 (ru) | 2018-10-09 | 2020-05-18 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для маршрутизации и выполнения транзакций |
RU2711348C1 (ru) | 2018-10-15 | 2020-01-16 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для обработки запросов в распределенной базе данных |
RU2714373C1 (ru) | 2018-12-13 | 2020-02-14 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для планирования выполнения операций ввода/вывода |
RU2749649C2 (ru) | 2018-12-21 | 2021-06-16 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для планирования обработки операций ввода/вывода |
RU2720951C1 (ru) | 2018-12-29 | 2020-05-15 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и распределенная компьютерная система для обработки данных |
RU2746042C1 (ru) | 2019-02-06 | 2021-04-06 | Общество С Ограниченной Ответственностью "Яндекс" | Способ и система для передачи сообщения |
KR102289140B1 (ko) * | 2019-12-03 | 2021-08-13 | 현대오트론 주식회사 | 전자 장치에서 소프트웨어를 실행하는 방법 및 그 전자 장치 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11224204A (ja) * | 1998-02-06 | 1999-08-17 | Nec Corp | 排他制御装置及びその方法 |
US20020174218A1 (en) * | 2001-05-18 | 2002-11-21 | Dick Kevin Stewart | System, method and computer program product for analyzing data from network-based structured message stream |
WO2008038195A2 (en) * | 2006-09-25 | 2008-04-03 | Koninklijke Philips Electronics N.V. | Robust critical section design in multithreaded applications |
JP2008146674A (ja) * | 2008-02-04 | 2008-06-26 | Toshiba Corp | 自動分散処理システム |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR0146551B1 (ko) * | 1995-08-21 | 1998-09-15 | 양승택 | 임계영역을 지원하는 래치 관리방법 |
US5828881A (en) * | 1995-11-09 | 1998-10-27 | Chromatic Research, Inc. | System and method for stack-based processing of multiple real-time audio tasks |
JP4197571B2 (ja) * | 2000-03-28 | 2008-12-17 | 株式会社東芝 | 自動分散処理システムおよび記録媒体 |
US6772153B1 (en) * | 2000-08-11 | 2004-08-03 | International Business Machines Corporation | Method and apparatus to provide concurrency control over objects without atomic operations on non-shared objects |
US7673304B2 (en) * | 2003-02-18 | 2010-03-02 | Microsoft Corporation | Multithreaded kernel for graphics processing unit |
US7290086B2 (en) * | 2003-05-28 | 2007-10-30 | International Business Machines Corporation | Method, apparatus and program storage device for providing asynchronous status messaging in a data storage system |
US7644197B1 (en) * | 2003-10-15 | 2010-01-05 | Sun Microsystems, Inc. | Queue management by multiple processors |
US7315912B2 (en) * | 2004-04-01 | 2008-01-01 | Nvidia Corporation | Deadlock avoidance in a bus fabric |
US7975272B2 (en) * | 2006-12-30 | 2011-07-05 | Intel Corporation | Thread queuing method and apparatus |
US8869169B2 (en) * | 2007-04-30 | 2014-10-21 | Accenture Global Services Limited | Alternately processing messages |
US8161460B2 (en) * | 2007-05-30 | 2012-04-17 | Sap Ag | System and method of automatic interoperability checking for BPEL |
-
2009
- 2009-10-30 KR KR1020090104500A patent/KR101080845B1/ko active IP Right Grant
-
2010
- 2010-10-29 CN CN201080049494.2A patent/CN102630315B/zh active Active
- 2010-10-29 US US13/505,170 patent/US8826301B2/en active Active
- 2010-10-29 WO PCT/KR2010/007517 patent/WO2011053038A2/ko active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11224204A (ja) * | 1998-02-06 | 1999-08-17 | Nec Corp | 排他制御装置及びその方法 |
US20020174218A1 (en) * | 2001-05-18 | 2002-11-21 | Dick Kevin Stewart | System, method and computer program product for analyzing data from network-based structured message stream |
WO2008038195A2 (en) * | 2006-09-25 | 2008-04-03 | Koninklijke Philips Electronics N.V. | Robust critical section design in multithreaded applications |
JP2008146674A (ja) * | 2008-02-04 | 2008-06-26 | Toshiba Corp | 自動分散処理システム |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021000656A1 (zh) * | 2019-07-04 | 2021-01-07 | 中兴通讯股份有限公司 | 死锁检测控制方法、装置、通信设备及计算机存储介质 |
Also Published As
Publication number | Publication date |
---|---|
US8826301B2 (en) | 2014-09-02 |
CN102630315A (zh) | 2012-08-08 |
CN102630315B (zh) | 2015-06-24 |
KR20110047753A (ko) | 2011-05-09 |
WO2011053038A3 (ko) | 2011-11-03 |
US20120216215A1 (en) | 2012-08-23 |
KR101080845B1 (ko) | 2011-11-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2011053038A2 (ko) | 교착 상태의 방지를 위한 데이터 처리 방법 및 시스템 | |
US7895260B2 (en) | Processing data access requests among a plurality of compute nodes | |
US7984450B2 (en) | Dispatching packets on a global combining network of a parallel computer | |
US7948999B2 (en) | Signaling completion of a message transfer from an origin compute node to a target compute node | |
KR101951072B1 (ko) | 코어 간 통신 장치 및 방법 | |
US7797292B2 (en) | Apparatus, system, and method for an alternate lock facility connection path | |
US8325633B2 (en) | Remote direct memory access | |
US7797445B2 (en) | Dynamic network link selection for transmitting a message between compute nodes of a parallel computer | |
JP3807250B2 (ja) | クラスタシステム、コンピュータ及びプログラム | |
WO2014025145A1 (en) | Method and apparatus for processing message between processors | |
US20070242611A1 (en) | Computer Hardware Fault Diagnosis | |
US7779173B2 (en) | Direct memory access transfer completion notification | |
US20080288691A1 (en) | Method and apparatus of lock transactions processing in single or multi-core processor | |
WO2016064158A1 (ko) | 재구성 가능 프로세서 및 그 동작 방법 | |
US6175931B1 (en) | Global hard error distribution using the SCI interconnect | |
WO2015023144A1 (ko) | 메모리 공유 환경에서 데이터 무결성 감시 장치 및 방법 | |
US7889657B2 (en) | Signaling completion of a message transfer from an origin compute node to a target compute node | |
US7006442B1 (en) | Communication control system and control method thereof | |
KR20050080704A (ko) | 프로세서간 데이터 전송 장치 및 방법 | |
WO2012118268A2 (ko) | 멀티 버추얼 머신을 이용한 멀티 쓰레드 처리시스템 및 그 방법 | |
WO2022114272A1 (ko) | 적응형 트랜잭션 처리 방법 및 이를 위한 장치 | |
US9015717B2 (en) | Method for processing tasks in parallel and selecting a network for communication | |
KR100978083B1 (ko) | 공유 메모리형 멀티 프로세서에 있어서의 절차 호출 방법 및 절차 호출 프로그램을 기록한 컴퓨터로 판독 가능한 기록 매체 | |
WO2018080192A1 (en) | Electronic device and data transfer method thereof | |
WO2023121104A1 (ko) | 리눅스 커널의 인터럽트 처리 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080049494.2 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10827122 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13505170 Country of ref document: US |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 06/08/2012) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 10827122 Country of ref document: EP Kind code of ref document: A2 |