Patent Application
20160077882
The present invention relates to a scheduling system, etc. that perform scheduling.
A space division method is a scheduling method used in a multiprocessor system when a plurality of independent tasks is processed. Referring to
Referring to
The task scheduler 45 receives a task to be executed as an input. Then, the task scheduler 45 reserves, by referencing the number of processors required for executing the received task and information about usage status of a plurality of processors (the processors 41 to 44) held by the server resource management unit 46, a processor (or processors) required for the execution. Then, the task scheduler 45 updates information held by the server resource management unit 46 and puts the task to the server 40. The task scheduler 45 updates the information held by the server resource management unit 46 after detecting completion of task execution by the server 40. The task scheduler 45 releases the processor reserved for processing the task.
The task scheduler 45 uses the processors (the processors 41 to 44) included in the server 40 for processing a plurality of tasks in accordance with the aforementioned operation. Thus, processing performance in the server 40 improves.
On the other hand, as illustrated in
Referring to
Referring to
Referring to
Referring to
A program execution control method disclosed in PTL 1 represents a method for power-saving control in a system including different types of processors. In other words, the execution control method is a control method for performance improvement. In accordance with the execution control method, a clock frequency is changed so that respective processors complete split tasks simultaneously.
A data processing apparatus disclosed in PTL 2 reduces overhead required for saving and restoration, depending on progress status of the interrupted process when interrupting a process in data processing to give priority to another process.
In a data processing apparatus disclosed by PTL 3, software executed on a processor and hardware dedicated to a specific process are carried out in order of priority. The data processing apparatus enhances processing efficiency related to task switching.
[PTL 1] Japanese Laid-open Patent Application No. 2011-197803
[PTL 2] Japanese Laid-open Patent Application No. 2010-181989
[PTL 3] Japanese Laid-open Patent Application No. 2007-102399
Referring to
A task scheduler 52 allocates a task to a resource by managing a resource in the many-core accelerator 49 when putting a task to the server 47. The task scheduler 52 releases the allocated resource when completing the task. In
Further, the task scheduler 52 reserves a maximum resource for processing the series of tasks on task activation even if a resource of the many-core accelerator 49 required for processing a series of tasks changes. Therefore, when a specific task using just part of a resource is processed in the series of tasks, there is a redundant resource that does not perform a process in Step S45. A problem that, as described above, there is a redundant resource that does not perform a specific process when a series of tasks are processed is referred to as an unused resource problem.
On the other hand, there exists a method in which, in order to avoid the aforementioned unused resource problem, the task scheduler 52 recognizes that the many-core accelerator 49 holds a more resource than the actual resource. However, as a result of avoiding the unused resource problem by such a method as described above, a resource of the many-core accelerator 49 becomes insufficient for actually processing a task. Therefore, the many-core accelerator 49 fails to process the task, or task processing becomes an excessively heavy load in the many-core accelerator 49. Consequently, processing performance possessed by a system 55 degrades.
In other words, a task scheduler 52 that adopts such a processing method as described above is not able to avoid the unused resource problem. Therefore, the many-core accelerator 49 degrades processing performance or fails task processing.
A main objective of the present invention is to provide a scheduling system, etc. more efficiently enabling processing performance possessed by a resource to be exhibited.
In order to achieve the object mentioned above, a scheduling system includes the following configuration.
In other word, a scheduling system including:
a scheduler configured to reserve a second communication channel, that is capable of transmitting/receiving a first data processed by a task between the memory and the accelerator memory, as a second communication resource in accordance with a fifth instruction for reserving the second communication channel from a first communication channel that is capable of transmitting/receiving data between a many-core accelerator to be a resource and a processor which controls the resource, and determine a specific resource for processing the task by referring to the first data transmitted/received via the second communication resource in accordance with a first instruction for reserving the resource; wherein
the task is processed by a calculation processing apparatus which includes the many-core accelerator, an accelerator memory accessed by the many-core accelerator, the processor, a memory accessed by the processor, and a first communication channel.
Also, as another aspect of the present invention, a scheduling method includes:
reserving a second communication channel, that is capable of transmitting/receiving a first data processed by a task between the memory and the accelerator memory, as a second communication resource in accordance with a fifth instruction for reserving the second communication channel from a first communication channel that is capable of transmitting/receiving data between a many-core accelerator to be a resource and a processor which controls the resource; and
determining a specific resource for processing the task by referring to the first data transmitted/received via the second communication resource in accordance with a first instruction for reserving the resource; wherein
the task is processed by a calculation processing apparatus which includes the many-core accelerator, an accelerator memory accessed by the many-core accelerator, the processor, a memory accessed by the processor, and a first communication channel.
Furthermore, the object is also realized by a scheduling program, and a computer-readable recording medium which records the scheduling program.
A scheduling system, etc. according to the present invention is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, exemplary embodiments of the present invention will be described in detail with reference to drawings.
A configuration included in a scheduling system 1 according to a first exemplary embodiment of the present invention and processes performed by the scheduling system 1 will be described in detail referring to
Referring to
The host processor 4 performs processing such as control related to the many-core accelerator 5. First, the host processor 4 starts processing of the task 6. The host processor 4 reads an instruction (also referred to as command; hereinafter an instruction for reserving a resource is also referred to as “first instruction”) to reserve a resource (many-core accelerator 5) from the task 6. Then, the host processor 4 transmits a command for reserving a resource to the scheduling system 1 in accordance with the read first instruction (Step S1).
Next, the scheduler 2 checks whether or not a resource (hereinafter abbreviated as “resource reservation”) for the task 6 is allocatable when receiving the command (Step S2). The scheduler 2 reserves a resource (Step S3) when resource a resource is decided to be allocatable (YES in Step S2). The scheduler 2 checks again whether or not resource reservation is possible (Step S2) when resource is decided not to be allocatable (NO in Step S2), When the scheduler 2 decides that resource is allocatable (YES in Step S2), the many-core accelerator 5 executes the task 6 (Step S4).
When resource is decided not to be allocatable (NO in Step S2), the scheduler 2 waits for the resource to be released by performing the aforementioned process. Then, the scheduler 2 releases the reserved resource (Step S5).
The scheduling system 1 may be realized as, for example, a function in an operating system. The scheduling system 1 may also, for example, perform such a process as described above by transmitting/receiving a parameter, etc. related to a resource to/from an operating system.
As described in “BACKGROUND ART” a system described in PTL 1 to PTL 3 continues reserving a maximum resource that processes a series of tasks during a period between start and end of task processing. Therefore, when processing of a series of tasks uses only part of a resource, some part of the resource does not perform processing.
On the other hand, the scheduling system 1 according to the first exemplary embodiment reserves a resource in accordance with a request from a task, and then the reserved resource performs processing. The scheduling system 1 releases the resource by the host processor 4 commanding release of the resource. Even when the server 3 processes a series of tasks, the scheduling system 1 is able to allocate a resource for processing each task depending on task processing. Therefore, the scheduling system 1 according to the first exemplary embodiment is capable of, even when a series of tasks are processed, alleviating a situation in which only part of a resource performs processing.
In other words, the scheduling system 1 according to the first exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, a second exemplary embodiment based on the aforementioned first exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned first exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
Since the management unit 9 manages information about usage status of a resource, the scheduler 8 is able to decide whether or not a resource can be allocated without referencing the outside. Therefore, the scheduling system 7 according to the second exemplary embodiment provides efficient management of a resource. Further, since the second exemplary embodiment includes a similar configuration to the first exemplary embodiment, the second exemplary embodiment can enjoy the similar effect to the first exemplary embodiment.
In other words, the scheduling system 7 according to the second exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, a third exemplary embodiment based on the aforementioned first exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned first exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
A host processor 4 executes the first part to be processed by the host processor 4 in the task 12 (Step S7). Then, the host processor 4 transmits a command for reserving a resource to the scheduler 11 in accordance with a first instruction (Step S8). When a resource is decided to be allocatable (YES in Step S9), the scheduler 11 reserves a resource (Step S10). When a resource is decided not to be allocatable (NO in Step S9), the scheduler 11 decides again whether or not a resource is allocatable (Step S9).
Next, the resource (included in the many-core accelerator 5) reserved by the scheduler 11 executes the second part to be processed by the resource (Step S11). Then, the host processor 4 issues a command for releasing the resource to the scheduler 11 in response to receiving a command for releasing the resource reserved by the scheduler 11 (hereinafter this command is referred to as “second instruction”) (Step S12). The scheduler 11 releases the reserved resource (Step S13) in response to receiving the command.
The first instruction includes, for example, information about the number of processors, etc. While the scheduler 11 determines an amount of resource on the basis of the aforementioned number of processors, etc., the amount of resource does not necessarily need to be equivalent to the aforementioned value. Further, the scheduler 11 may transmit information about a reserved resource. The information about the reserved resource may include information about the number of reserved processors, a list of available processor numbers, etc.
The task 12 includes the first part processed by the host processor 4, the second part, and the first instruction that reserves a resource for execution of the second part. Therefore, the scheduler 11 reserves a required resource before processing of the second part, and releases the resource after the reserved resource completes processing of the second part. In other words, the scheduling system 10 according to the third exemplary embodiment provides more detailed resource management compared with a system disclosed in PTL 1 to 3.
In other words, the scheduling system 10 according to the third exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
While, for convenience of description, the third exemplary embodiment is based on the first exemplary embodiment in the aforementioned description, the third exemplary embodiment may also be based on the second exemplary embodiment. In that case, the third exemplary embodiment can enjoy the similar effect to the second exemplary embodiment.
Next, a fourth exemplary embodiment based on the aforementioned first exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned first exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
The server 16 includes a host processor 18, a main memory 19 that stores data processed by the host processor 18, a many-core accelerator 17, and an accelerator memory 20 that stores data processed by the many-core accelerator 17.
The scheduling system 13 includes a scheduler 14. The task 15 includes, in addition to the aforementioned first part, first instruction, and second part, a third part that is an instruction for transmitting data from the main memory 19 to the accelerator memory 20 and a fourth part that is an instruction for transmitting data from the accelerator memory 20 to the main memory 19.
In accordance with the first instruction after executing the first part, the host processor 18 transmits a request for reserving a specific resource to the scheduling system 13 (Step S14). The scheduler 14 reserves a specific resource after receiving the request (Step S15). Step S15 is a collective expression including a series of processes in Steps S2 and S3 in
Next, the specific resource reserved by the scheduler 14 executes the second part (Step S17). Then, the host processor 18 transmits data processed by the specific resource from the accelerator memory 20 to the main memory 19 (Step S18). Then, the host processor 18 transmits a request for releasing the specific resource to the scheduling system 13 in accordance with the second instruction (Step S19). Then, the scheduler 14 releases the specific resource after receiving the request (Step S20).
In the fourth exemplary embodiment, the scheduler 14 may reserve the accelerator memory 20 in addition to a processing apparatus in the many-core accelerator 17. In this case, a specific many-core accelerator 17 refers a specific accelerator memory 20. Referring to
After executing the first part, the host processor 18 transmits a request for reserving a specific accelerator memory 20 to the scheduling system 13 in accordance with the first instruction (Step S30). The scheduler 14 reserves a specific accelerator memory 20 in response to receiving the request (Step S31). Then, the host processor 18 transmits data processed by the many-core accelerator 17 from the main memory 19 to the specific accelerator memory 20 (Step S16).
Next, the host processor 18 makes a request for reserving a specific resource to the scheduling system 13 (Step S14). The scheduler 14 reserves a specific resource after receiving the request (Step S15). Step S15 is a collective expression including a series of processes in Steps S2 and S3 in
Next, the specific resource reserved by the scheduler 14 executes the second part (Step S17). Then, the host processor 18 transmits a request for releasing the specific resource to the scheduling system 13 in accordance with the second instruction (Step S19). Then, the scheduler 14 releases the specific resource in response to receiving the request (Step S20). Then, the host processor 18 transmits data processed by the specific resource from the accelerator memory 20 to the main memory 19 (Step S18).
Next, the host processor 18 transmits a request for releasing the specific accelerator memory 20 to the scheduler 14 (Step S32). Then, the scheduler 14 releases the specific accelerator memory 20 in response to receiving the request (Step S33).
The scheduling system 13 according to the fourth exemplary embodiment is also capable of efficiently managing data to be processed by the many-core accelerator 17 with a resource or the accelerator memory 20 in the system 57.
In other words, the scheduling system 13 according to the fourth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
While, for convenience of description, the fourth exemplary embodiment is based on the first exemplary embodiment in the aforementioned description, the fourth exemplary embodiment may be based on the second exemplary embodiment or the third exemplary embodiment. In that case, the fourth exemplary embodiment can enjoy the similar effect to the second or third exemplary embodiment.
Next, a fifth exemplary embodiment based on the aforementioned third exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned third exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
When the scheduler 22 decides that a resource cannot be allocated (NO in Step S9), the host processor 4 executes the fifth part (Step S21). When the scheduler 22 decides that a resource can be allocated (YES in Step S9), the scheduler 22 reserves a specific resource (Step S10).
The scheduling system 21 according to the fifth exemplary embodiment allows the host processor 4 to perform processing instead of the many-core accelerator 5 depending on resource status in the many-core accelerator 5. In other words, the task 23 can be processed more efficiently with the scheduling system 21 according to the fifth exemplary embodiment.
In other words, the scheduling system 21 according to the fifth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, a sixth exemplary embodiment based on the aforementioned first exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned first exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
The second task scheduler 26 transmits, for example, information related to a task such as the number of tasks in the task 6 to the scheduling system 24 (Step S23). Then, the scheduler 25 calculates an amount of resource on the basis of the received information (Step S24). For example, the scheduler 25 may calculate an amount of resource by dividing the number of logical processors included in the many-core accelerator 5 by the number of tasks put to the server 3 by the second task scheduler 26, or may calculate an amount of resource by multiplying the amount of the value calculated above by two. A method by which the scheduling system 24 calculates an amount of resource is not limited to the aforementioned example.
The scheduling system 24 receives information usable for resource allocation control from the second task scheduler 26. Thus, the scheduling system 24 is able to perform scheduling more efficiently and adjust a load to the many-core accelerator 5.
In other words, the scheduling system 24 according to the sixth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, a seventh exemplary embodiment based on the aforementioned second exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned second exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
Referring to
The scheduler 28 reads load information including a load value representing the load of a resource in the many-core accelerator 5 from the management unit 29 (Step S25). Then, the scheduler 28 compares a predetermined second threshold value with the read load value. When the read load value is decided to be less than the predetermined second threshold value, in other words, when the load status is decided to be low (YES in Step S26), the scheduler 28 transmits a signal requesting more tasks to be input to the second task scheduler 30 (Step S27). The scheduler 28 compares a predetermined first threshold value with the read load value. When the read load value is decided to be greater than the predetermined first threshold value, in other words, when the load status is decided to be high (NO in Step S26), the scheduler 28 transmits a signal requesting less tasks to be input to the second task scheduler 30 (Step S28).
Next, the second task scheduler 30 adjusts a task amount in accordance with the signal (Step S29).
Since the scheduling system 27 transmits a signal to the second task scheduler 30 regarding load information about a resource, the scheduling system 27 according to the seventh exemplary embodiment is able to adjust a load to the many-core accelerator 5 to an appropriate level.
In other words, the scheduling system 27 according to the seventh exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, an eighth exemplary embodiment based on the aforementioned fourth exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned fourth exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
A system 117 includes a server 100 and the scheduling system 114.
The server 100 includes a host processor 101, a main memory 103, a many-core accelerator 102, an accelerator memory 104, and a communication channel 105 that connects the host processor 101 and the many-core accelerator 102. The host processor 101 and the many-core accelerator 102 communicate (also referred to as access or transmit/receive) data to be referenced via the communication channel 105.
The scheduling system 114 includes a scheduler 115 and a communication channel scheduler 116 that reserves a communication resource included in the communication channel 105.
First, the host processor 101 sends a first instruction 201 that reserves a resource in the many-core accelerator 102 to the scheduling system 114 (Step S14). The scheduling system 114 reserves a specific resource in the many-core accelerator 102 in accordance with the first instruction 201 (Step S15). Then, the host processor 101 receives a fifth instruction 110 for reserving the communication channel 105. The fifth instruction 110 is included in a task 106. In accordance with the received fifth instruction 110, the host processor 101 commands the communication channel scheduler 116 to reserve a communication resource included in the communication channel 105 (Step S101).
Next, the communication channel scheduler 116 receives a command for reserving a communication resource capable of communicating traffic based on the fifth instruction 110 from the host processor 101. The communication channel scheduler 116 measures traffic on the communication channel 105 until the communication channel 105 becomes capable of communicating traffic based on the fifth instruction 110. When the communication channel 105 is capable of communicating traffic based on the fifth instruction 110, the communication channel scheduler 116 reserves a communication resource based on the fifth instruction 110 from the communication channel 105 (Step S102).
Next, the host processor 101 transmits data accessed by the many-core accelerator 102 from the main memory 103 to the accelerator memory 104 via the communication resource reserved by the communication channel scheduler 116 (Step S16).
In other words, the host processor 101 transmits data from the main memory 103 to the accelerator memory 104. Then, the host processor 101 transmits a request for releasing the reserved communication resource to the communication channel scheduler 116 in accordance with a sixth instruction 111 for releasing the reserved communication resource (Step S103).
The communication channel scheduler 116 receives a request for releasing the reserved communication resource and releases the reserved communication resource in accordance with the received request (Step S104).
Next, a specific resource reserved by the scheduler 115 executes a second part 107 in the task 106 by accessing data stored in the accelerator memory 104 (Step S17).
The host processor 101 transmits a request for reserving a communication resource capable of communicating traffic based on the seventh instruction 112 to the communication channel scheduler 116 in accordance with a seventh instruction 112 for reserving a communication resource included in the communication channel 105 (Step S105).
The communication channel scheduler 116 receives a command for reserving a communication resource from the host processor 101. Then, the communication channel scheduler 116 measures traffic on the communication channel 105 until a communication resource capable of communicating traffic designated by the seventh instruction 112 can be reserved. When the communication channel 105 is capable of communicating traffic based on the seventh instruction 112, the communication channel scheduler 116 reserves a communication resource based on the seventh instruction 112 from the communication channel 105 (Step S106).
The host processor 101 transmits data processed by the specific resource, etc. from the accelerator memory 104 to the main memory 103 via the communication resource reserved by the communication channel scheduler 116 (Step S18).
The host processor 101 receives an eighth instruction 113 for releasing the reserved communication resource from the task 106. In accordance with the received eighth instruction 113, the host processor 101 transmits a command for releasing the reserved communication resource to the communication channel scheduler 116 (Step S107).
The communication channel scheduler 116 receives the command from the host processor 101 and releases the reserved communication resource in accordance with the received command (Step S108).
In accordance with the aforementioned second instruction 202, the host processor 101 transmits a command for releasing the specific resource to the scheduling system 114 (Step S19).
The scheduling system 114 receives the command from the host processor 101 and releases the reserved specific resource in accordance with the received command (Step S20).
In the aforementioned example, it was assumed that the fifth instruction 110, the sixth instruction 111, the seventh instruction 112, and the eighth instruction 113 were described to represent instructions for reserving a communication resource or instructions for releasing a communication resource. However, the instructions may include other information. For example, the fifth instruction to the eighth instruction may include information identifying a task 106 issuing the instruction or may include time of receipt of the instruction. In addition, the fifth instruction to the eighth instruction may include a size of data transmitted/received between the main memory 103 and the accelerator memory 104 or may include information about a data structure in the main memory 103, or the like.
In the aforementioned example, it was assumed that the fifth instruction 110 and the seventh instruction 112 were described to be different instructions. However, the instructions may be a same instruction. Similarly, in the aforementioned example, it was assumed that the sixth instruction 111 and the eighth instruction 113 were described to be different instructions. However, the instructions may be a same instruction. In this case, the task 106 executes the fifth instruction 110 instead of the seventh instruction 112 and executes the eighth instruction 113 instead of the sixth instruction 111.
In the case of the aforementioned example, the communication channel scheduler 116, for example, reserves, in accordance with the seventh instruction 112, a process related to the fifth instruction 110, specifically, a communication resource. Similarly, the communication channel scheduler 116 releases a process related to the sixth instruction 111, specifically, a communication resource in accordance with the eighth instruction 113.
In addition, a third part 108 may include a process that reserves a storage area in the accelerator memory 104. Similarly, a fourth part 109 may include a process that releases a storage area in the accelerator memory 104.
Further, in the present exemplary embodiment, the scheduler 115 may reserve a storage area in the accelerator memory 104. In this case, the scheduling system 114 performs processes illustrated in
First, the host processor 101 transmits a command for reserving a certain size of a storage area in the accelerator memory 104 to the scheduling system 114 (Step S30). Then, the scheduler 115 receives the command from the host processor 101 and reserves a certain size of a storage area in the accelerator memory 104 in accordance with the received command (Step S31).
Next, the host processor 101 receives the fifth instruction 110 for reserving the communication channel 105 in the task 106 and transmits a command for reserving a communication resource included in the communication channel 105 to the communication channel scheduler 116 in accordance with the received fifth instruction 110 (Step S101).
Next, the communication channel scheduler 116 receives the command for reserving a communication resource based on the fifth instruction 110 from the host processor 101. The communication channel scheduler 116 measures traffic (hereinafter “communication band” is used as a synonym) transmitted/received on the communication channel 105 until the communication channel 105 becomes capable of communicating traffic based on the fifth instruction 110. When the communication channel 105 is capable of communicating traffic based on the fifth instruction 110, the communication channel scheduler 116 reserves a communication resource based on the fifth instruction 110 (Step S102).
Next, the host processor 101 transmits data processed by the many-core accelerator 102 from the main memory 103 to a specific storage area in the accelerator memory 104 via the communication resource reserved by the communication channel scheduler 116 (Step S16).
The host processor 101 transmits a command for releasing the reserved communication resource to the communication channel scheduler 116 in accordance with the sixth instruction 111 that releases the reserved communication resource (Step S103).
The communication channel scheduler 116 receives the command for releasing the reserved communication resource and releases the reserved communication resource in accordance with the received command (Step S104).
Next, the host processor 101 transmits a command for reserving a resource to the scheduling system 114 in accordance with the first instruction 201 that reserves a resource in the many-core accelerator 102 (Step S14). The scheduler 14 reserves a specific resource in the many-core accelerator 102 in response to receiving the command (Step S15).
The specific resource reserved by the scheduler executes the second part 107 in the task 106 by processing data transmitted to the accelerator memory 104 (Step S17).
After the specific resource completes processing in the second part 107, the host processor 101 commands the scheduling system 114 to release the specific resource in accordance with the second instruction 202 that releases the reserved resource (Step S19). Then, the scheduler 14 releases the specific resource in accordance with the command (Step S20).
In the task 106, the host processor 101 transmits a command for reserving a communication resource based on the seventh instruction 112 to the communication channel scheduler 116 in accordance with the seventh instruction 112 that reserves a communication resource included in the communication channel 105 (Step S105).
The communication channel scheduler 116 receives a command for reserving a communication resource from the host processor 101. Then, the communication channel scheduler 116 measures traffic transmitted/received on the communication channel 105 until the communication channel 105 becomes capable of communicating traffic based on the seventh instruction 112. When the communication channel 105 is capable of communicating traffic based on the seventh instruction 112, the communication channel scheduler 116 reserves a communication resource based on the seventh instruction 112 (Step S106).
The host processor 101 transmits data processed by the specific resource, etc. from the accelerator memory 104 to the main memory 103 via the communication resource reserved by the communication channel scheduler 116 (Step S18).
The host processor 101 receives the eighth instruction 113 for releasing the reserved communication resource from the task 106 and commands the communication channel scheduler 116 to release the reserved communication resource in accordance with the received eighth instruction 113 (Step S107).
The communication channel scheduler 116 receives the command from the host processor 101 and releases the reserved communication resource in accordance with the received command (Step S108).
The host processor 101 transmits a command for releasing the reserved specific storage area to the scheduler (Step S32). Then, the scheduler 14 releases the reserved specific storage area in response to receiving the command (Step S33).
Since the eighth exemplary embodiment includes a similar configuration to the fourth exemplary embodiment, the eighth exemplary embodiment can enjoy the similar effect to the fourth exemplary embodiment. In other words, the scheduling system 114 according to the eighth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
In the present exemplary embodiment, the communication channel scheduler 116 reserves a communication resource in accordance with an instruction in the task 106. The host processor 101 and the many-core accelerator 102 transmit/receive data between the main memory 103 and the accelerator memory 104 via the communication resource. Therefore, the present exemplary embodiment reduces possibility of communication delay on the communication channel 105. Further, when the communication channel scheduler 116 is not able to reserve an amount of a communication resource capable of transmitting/receiving data based on an instruction in the task 106, data transmission/reception required for processing the task 106 is temporarily halted in advance. Consequently, according to the present exemplary embodiment, possibility of data transmission/reception for a task other than the task being interfered is low.
In other words, according to the present exemplary embodiment, the communication channel scheduler 116 controls communication transmitted/received between the main memory 103 and the accelerator memory 104 based on communication performance of the communication channel 105. Therefore, an amount of data to be transmitted between the main memory 103 and the accelerator memory 104 per unit time is less than an amount of transmission that can be transmitted per unit time on the communication channel. Therefore, possibility of delay in processing in the many-core accelerator 102 due to communication delay on the communication channel 105 is low. Consequently, the present exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
In the present exemplary embodiment, it was assumed that the scheduler 115 was described to include the communication channel scheduler 116 for convenience of description. However, the scheduler 115 may realize a function in the communication channel scheduler 116.
Next, a ninth exemplary embodiment based on the aforementioned eighth exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned eighth exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
A system 255 includes a server 100 and the scheduling system 251.
The scheduling system 251 includes a scheduler 115 and a communication channel scheduler 252. Further, the communication channel scheduler 252 includes a communication information unit 254 and a communication control unit 253.
First, a host processor 101 transmits a command for reserving a communication resource to the communication channel scheduler 252 in accordance with the fifth instruction 110 in Step S101 or the seventh instruction 112 in Step S105.
Next, the communication channel scheduler 252 receives a command for reserving a communication resource from the host processor 101 (Step S201). Then, the communication control unit 253 in the communication channel scheduler 252 examines whether or not a communication channel 105 includes a communication channel in an unused (also referred to as “dormant,” “idle,” “standby,” etc.) state. An “unused state” represents a state in which a target apparatus is not assigned to a task, etc.
For example, the communication control unit 253 examines whether or not the communication channel 105 includes a communication channel in an unused state on the basis of a communication channel usage representing the total number of communication channels used by a task. In accordance with the received command, the communication control unit 253 compares a calculated value obtained by adding 1 to the communication channel usage with the number of communication channels originally included in the communication channel 105 in accordance with the received command (Step S202).
When the calculated communication channel usage is equal to or less than the number of communication channels included in the communication channel 105 (YES in Step S202), the communication control unit 253 updates the communication channel usage to the calculated value (Step S203).
The communication channel scheduler 252 reserves a communication resource from a communication channel in an unused state in accordance with the received command (Step S204).
When the calculated communication channel usage is greater than the number of communication channels included in the communication channel 105 (NO in Step S202), the communication control unit 253 stores a task identifier into the communication information unit 254 (Step S205). In other words, as illustrated in
For example, the communication information unit 254 stores a task identifier “1,” a task identifier “3,” a task identifier “4,” and a task identifier “2.” In this case, the communication channel scheduler 252 does not reserve a communication resource for tasks with the task identifier “1,” the task identifier “3,” the task identifier “4,” and the task identifier “2” in the process in Step S205.
Next, processes performed when the host processor 101 commands the communication channel scheduler 252 to release a reserved communication resource in accordance with the sixth instruction 111 in Step S103 or the eighth instruction 113 in Step S107 will be described.
First, the host processor 101 receives a sixth instruction 111 or an eighth instruction 113 from the task 106. Then, the host processor 101 transmits a command for releasing a reserved communication resource to the communication channel scheduler 252 in accordance with the received sixth instruction 111 or the eighth instruction 113. Then, the communication channel scheduler 252 receives a command for releasing the reserved communication resource from the host processor 101 (Step S211).
Next, the communication control unit 253 in the communication channel scheduler 252 decides whether or not the communication information unit 254 stores a task identifier (Step S212). When the communication information unit 254 is decided to store a task identifier (YES in Step S212), the communication control unit 253 reads a specific task identifier from the communication information unit 254 (Step S213). Then, the communication control unit 253 controls communication for the task represented by the read task identifier as illustrated in Steps S202 to S205 in
Next, the communication channel scheduler 252 subtracts 1 from the communication channel usage and sets the communication channel usage to the calculated value (Step S215). The communication channel scheduler 252 releases the reserved communication resource in accordance with the command for releasing a communication resource (Step S216). The communication channel scheduler 252 may process Step S216 prior to Step S215.
In the aforementioned example, a function implemented by use of the communication channel usage may be implemented with a similar value such as the number of communication channels in an unused state.
Further, the communication information unit 254 may include a queue structure capable of storing task identifiers. In other words, the communication information unit 254 may include a data structure capable of storing task identifiers in order of time of receipt and outputting the stored task identifiers in order of time of receipt. In this case, the communication control unit 253 performs the processes illustrated in Step S212 or later, in order of time of receiving a command for reserving a communication resource.
Since the ninth exemplary embodiment includes a similar configuration to the eighth exemplary embodiment, the ninth exemplary embodiment can enjoy the similar effect to the eighth exemplary embodiment. In other words, the scheduling system 251 according to the ninth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Further, the communication channel scheduler 252 allocates a communication resource to a task 106, to which a communication resource is not allocatable in response to a command for reserving the communication channel 105, depending on a process of releasing a communication resource for another task. Therefore, the present exemplary embodiment provides more efficient allocation of a communication resource. Consequently, more efficient access is provided between the accelerator memory 104 and the main memory 103. In other words, the scheduling system 251 according to the ninth exemplary embodiment provides further reduction of processing performance degradation.
In the aforementioned example, it was assumed that the communication channel scheduler 252 managed a communication resource in the communication channel 105 based on a communication channel usage. However, the communication channel scheduler 252 may manage a communication resource by a communication channel measurement function that measures a communication band used in the communication channel 105.
In this example, the communication channel scheduler 252 compares a communication band measured by the communication channel measurement function with an available communication band in the communication channel 105. When the measured communication band is less than the available communication band, the communication channel scheduler 252 reserves a communication resource in accordance with a request for reserving the communication channel 105 (Step S204 in
Next, a tenth exemplary embodiment based on the aforementioned ninth exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned ninth exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
A system 306 includes a server 100 and the scheduling system 301.
The scheduling system 301 includes a scheduler 115 and a communication channel scheduler 302. The communication channel scheduler 302 includes the communication control unit 305, a communication information unit 304, and the priority order setting unit 303.
The communication information unit 304 is able to store, as illustrated in
For example, the first row in
The first data and the second data are respectively associated with a priority order “1” and a priority order “2”. Since the priority order of the first data is smaller than the priority order of the second data, the communication control unit 305 processes the first data prior to the second data.
The priority order setting unit 303 decides whether or not data stored in the communication information unit 304 are updated (Step S261). For example, when the communication control unit 305 updates a value in the communication information unit 304 in accordance with the fifth instruction 110 or the seventh instruction 112, the priority order setting unit 303 decides that data stored in the communication information unit 304 are updated.
When data stored in the communication information unit 304 are decided to be updated (YES in Step S261), the communication control unit 305 assigns the aforementioned priority order to the task 106 depending on the instruction type, etc. of the task 106 (Step S262). In this case, the communication control unit 305 assigns a priority order in accordance with a predetermined priority order assignment method. Further, when data stored in the communication information unit 304 are decided not to be updated (NO in Step S261), the communication control unit 305 does not perform the aforementioned process.
For example, as a predetermined priority order assignment method, there is a method in which a task with an instruction type “5” is assigned a higher priority order than a task with an instruction type “7”. In this case, the communication control unit 305 processes a task with the instruction type “5” in preference to a task with the instruction type “7”. The instruction type “7” represents the seventh instruction. Similarly, the instruction type “5” represents the fifth instruction.
The communication control unit 305 receives a command for releasing a reserved communication resource from the host processor 101 in accordance with a sixth instruction 111 or an eighth instruction 113 (Step S211). Then, the communication control unit 305 reads whether or not the communication information unit 304 stores a task identifier (Step S212). When the communication information unit 304 is decided not to store a task identifier (NO in Step S212), the communication control unit 305 releases the reserved communication resource (Step S216).
For example, when a communication resource in the communication channel 105 is managed based on a communication channel usage as the aforementioned example, the communication control unit 305 may calculate a value by subtracting 1 from the communication channel usage and update the communication channel usage with the calculated value (Step S215).
On the other hand, when the communication information unit 304 is decided to store a task identifier (YES in Step S212), the communication control unit 305 reads a task identifier associated with a high priority order from the communication information unit 304 (Step S313). For example, the communication control unit 305 may read a task identifier with the highest priority order in the communication information unit 304. Then, the communication control unit 305 performs processes such as reservation of a communication resource as illustrated in
Similar to the aforementioned processes, the communication control unit 305 may calculate a value by subtracting 1 from a communication channel usage and update the communication channel usage with the calculated value (Step S215).
A predetermined priority order assignment method includes, for example, assignment methods illustrated in
For example, a task with an instruction type “5” associated with the task identifier is assigned a higher priority order than a task with an instruction type “7” in an examples (
After reserving a communication resource in accordance with the fifth instruction 110, the scheduling system 301 assigns processing of the second part 107 to the many-core accelerator 102. The scheduling system 301 reserves a communication resource in accordance with the seventh instruction 112 after the many-core accelerator 102 completes processing of the second part 107. The scheduling system 301 processes a task 106 commanding the fifth instruction 110 in preference to a task 106 commanding the seventh instruction 112. Thus, the many-core accelerator 102 is able to transmit/receive data related to the second part 107 early via a communication resource. Consequently, processing efficiency in the many-core accelerator 102 becomes yet higher compared with the ninth exemplary embodiment.
For example, a predetermined priority order assignment method illustrated in
In the predetermined priority order assignment method illustrated in
Further, for example, a predetermined priority order assignment method illustrated in
In other words, the predetermined priority order assignment method assigns a higher priority order to a task 106 with a smaller size of the data among tasks associated with an instruction type “5” and a same time (Step S274). Further, the predetermined priority order assignment method assigns a higher priority order to a task with a smaller size of the data among tasks associated with an instruction type “7” and a same time (Step S275). The processes may be performed in the order of Steps S271, S273, S275, S272, and S274.
The present exemplary embodiment provides the task 106 with shorter waiting time in communication resource allocation.
When the scheduling system 301 allocates a communication resource to a first task, a second task in the communication information unit 304 waits for the communication resource to be released. As a size of data transmitted/received between the accelerator memory 104 and the main memory 103 becomes larger, it takes longer time to transmit/receive the data, thus a task waiting time for communication resource allocation becomes longer. According to the present exemplary embodiment, a task with a smaller size of data is assigned a higher priority order, therefore, the aforementioned waiting time is short.
Further, for example, a predetermined priority order assignment method illustrated in
Further, according to the method, when there is no storage area in an unused state in the accelerator memory 104 (NO in Step S281), a priority order is assigned (Step S282). In this case, the predetermined priority order assignment method is a method in which a task associated with the seventh instruction 112 is assigned a higher priority order than a task associated with the fifth instruction 110.
When there is no storage area in an unused state in the accelerator memory 104, the scheduling system 301 can allocate a communication resource in accordance with the fifth instruction 110. However, the host processor 101 is not able to transmit data required for processing in the second part 107 to the accelerator memory 104. In this case, the scheduling system 301 reserves a storage area in an unused state in the accelerator memory 104 and transmits the required data to the storage area in an unused state. Thus, the many-core accelerator 102 starts processing in the second part 107.
In other words, when there is no storage area in an unused state in the accelerator memory 104, the communication channel scheduler 302 assigns a high priority order to a task 106 associated with the seventh instruction 112. Thus, the communication channel scheduler 302 first reserves a storage area in an unused state in the accelerator memory 104. The host processor 101 transmits the referenced data from the main memory 103 to the storage area in an unused state. Then, the many-core accelerator 102 performs processing in the second part 107 on the basis of the data.
In other words, the communication channel scheduler 302 according to the present exemplary embodiment allocates a communication resource depending on usage status of the accelerator memory 104. Therefore, the communication channel scheduler 302 reduces possibility that, although there is no sufficient storage area to store data for processing a task in the accelerator memory 104, a communication resource for transmitting the data is allocated to the task. Thus, the communication channel scheduler 302 according to the present exemplary embodiment is able to reduce possibility of processing in the many-core accelerator 102 being halted due to the accelerator memory 104 not being able to store data to be processed by a task 106.
Since the tenth exemplary embodiment includes a similar configuration to the ninth exemplary embodiment, the tenth exemplary embodiment can enjoy the similar effect to the ninth exemplary embodiment. In other words, the scheduling system 301 according to the tenth exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Next, an eleventh exemplary embodiment based on the aforementioned fourth exemplary embodiment will be described.
In the following description, characteristic matters according to the present exemplary embodiment will be mainly described and an overlapping description of the same configuration as in the aforementioned fourth exemplary embodiment will be omitted by assigning the same reference sign thereto.
Referring to
A system 3006 includes a server 100 and the scheduling system 3004.
The scheduling system 3004 includes a scheduler 3005.
First, the scheduler 3005 controls the server 100 on the basis of a task 3001 including a first instruction 3002 for reserving a resource included in a many-core accelerator 102 and a fifth instruction 3003 for reserving a communication resource included in a communication channel 105.
A host processor 101 transmits a request for reserving a communication resource capable of communicating an amount instructed by the fifth instruction 3003 to the scheduler 3005 in accordance with the fifth instruction 3003 in the task 3001, (Step S3201).
Next, the scheduler 3005 decides whether or not there is a communication resource in an unused state in the communication channel 105 in response to receiving the request (Step S3202). When a communication resource capable of communicating the amount instructed by the fifth instruction is decided to be allocatable from the communication channel 105 (YES in Step S3202), the scheduler 3005 reserves the communication resource (Step S3203).
Next, the host processor 101 reads data to be processed by the task 3001 from a main memory 103 and transmits the read data to the many-core accelerator 102 (Step S3204). Then, the many-core accelerator 102 receives the data and stores the received data in an accelerator memory 104 (Step S3205).
Next, the host processor 101 transmits a request for reserving a resource for processing the task 3001 out of the many-core accelerator 102 to the scheduler 3005 in accordance with the first instruction 3002 (Step S3206).
The scheduler 3005 receives the request and decides whether or not a resource based on the first instruction 3002 can be reserved in the many-core accelerator 102 (Step S3207). When the many-core accelerator 102 is decided to include an unused-state resource to process the task 3001 (YES in Step S3207), the scheduler 3005 reserves a specific resource (Step S3208).
Next, the specific resource performs processing related to the task 3001 on the basis of the received data in the accelerator memory 104 (Step S3209).
Since the eleventh exemplary embodiment includes a similar configuration to the fourth exemplary embodiment, the eleventh exemplary embodiment can enjoy the similar effect to the fourth exemplary embodiment. In other words, the scheduling system 3004 according to the eleventh exemplary embodiment is capable of more efficiently enabling processing performance possessed by a resource to be exhibited.
Further, the scheduling system 3004 reserves a communication channel resource that transmits first data related to the task 3001 out of the communication channel 105. When the scheduler 3005 reserves a communication resource, the first data in the main memory 103 can be stored in the accelerator memory 104 without delay by the host processor. Consequently, a situation in which, although the scheduler 3005 reserves a specific resource, the specific resource is not able to perform processing related to the task 3001 due to delay of the first data, does not occur.
In other words, the scheduling system 3004 according to the present exemplary embodiment is capable of yet more efficiently enabling processing performance possessed by a resource to be exhibited in addition to the aforementioned effects.
A configuration example of a hardware resource realizing a scheduling system according to each exemplary embodiment of the present invention described above with a single calculation processing apparatus (information processing apparatus, computer) will be described. Such a scheduling system may be physically or functionally realized by use of at least two calculation processing apparatuses. Further, such a scheduling system may be realized as a dedicated apparatus.
The non-volatile recording medium 35 refers to, for example, a Compact Disc, a Digital Versatile Disc, a Blu-ray Disc (registered trademark), a Universal Serial Bus memory (USB memory), Solid State Drive, etc. that are computer-readable, capable of holding such a program without power supply, and portable. The non-volatile recording medium 35 is not limited to the aforementioned media. Further, such a program may be transferred via a communication network instead of the non-volatile recording medium 35.
When executing a software program (computer program, hereinafter simply referred to as “program”) stored in the disk 34, the CPU 32 copies the program to the memory 33 and performs arithmetic processing. The CPU 32 reads data required for executing the program from the memory 33. When a display is required, the CPU 32 displays an output result on the output apparatus 37. When inputting a program from outside, the CPU 32 reads the program from the input apparatus 36. The CPU 32 interprets and executes a scheduling program (processes performed by the scheduling system in
In such a case, the present invention can be regarded as realizable also by such a scheduling program. Further, the present invention can be regarded as realizable also by a computer-readable recording medium containing such a scheduling program.
The aforementioned respective exemplary embodiments may also be described in whole or part as the following Supplemental Notes. However, the present invention exemplified in each of the aforementioned exemplary embodiments is not limited to the following.
A scheduling system including a scheduler configured to reserve a second communication channel, that is capable of transmitting/receiving a first data processed by a task between the memory and the accelerator memory, as a second communication resource in accordance with a fifth instruction for reserving the second communication channel from a first communication channel that is capable of transmitting/receiving data between a many-core accelerator to be a resource and a processor which controls the resource, and determine a specific resource for processing the task by referring to the first data transmitted/received via the second communication resource in accordance with a first instruction for reserving the resource; wherein
the task is processed by a calculation processing apparatus which includes the many-core accelerator, an accelerator memory accessed by the many-core accelerator, the processor, a memory accessed by the processor, and a first communication channel.
The scheduling system according to Supplemental Note 1, wherein, based on the task including a first part processed by the processor, the fifth instruction, a third part that is an instruction for transmitting the first data from the memory to the accelerator memory via the second communication resource, a sixth instruction for releasing the second communication resource, a first instruction for reserving a resource processing the task from the resource, a second part that is an instruction for processing the task by the reserved resource, a seventh instruction for reserving, from the first communication channel, a third communication channel capable of transmitting/receiving second data transmitted from the accelerator memory to the memory, a fourth part that is an instruction for transmitting the second data from the accelerator memory to the memory via the third communication channel, an eighth instruction for releasing the third communication channel, and a second instruction for releasing the reserved resource processing the task, the scheduler includes a communication channel scheduler that reserves the second communication resource in accordance with the fifth instruction, releases the second communication resource in accordance with the sixth instruction, reserves the third communication channel as a third communication resource in accordance with the seventh instruction, and releases the third communication resource in accordance with the eighth instruction,
the scheduler reserves a resource for processing the task as a specific resource in accordance with the first instruction during execution of the first part by the processor and execution of the third part by the processor and releases the specific resource in accordance with the second instruction after execution of the fourth part by the processor,
the processor transmits the first data from the memory to the accelerator memory via the second communication resource and transmits the second data from the accelerator memory to the memory via the third communication resource, and
the specific resource generates the second data by processing the task while accessing the first data in accordance with the second part.
The scheduling system according to Supplemental Note 2, wherein the communication channel scheduler reserves a fourth communication resource from a communication channel in an unused state in the first communication channel in accordance with the fifth instruction or the seventh instruction, and does not reserve the fourth communication resource when the first communication channel does not include the communication channel in an unused state.
The scheduling system according to Supplemental Note 3, wherein
the task is associated with a task identifier that identifies a task, and
the communication channel scheduler includes:
communication information means capable of storing the task identifier,
communication control means for storing the task identifier associated with the task in the communication information means when the fourth communication resource cannot be reserved for the task executing the fifth instruction or the seventh instruction and performing communication control processing that reserves the fourth communication resource from the first communication channel in the unused state when the fourth communication resource can be reserved, and
releasing the fourth communication resource in accordance with the sixth instruction or the eighth instruction, reading the task identifier from the communication information means, and performing the communication control processing on the task associated with the read task identifier.
The scheduling system according to Supplemental Note 4, that includes
priority order setting means configured to, in accordance with a predetermined priority order assignment method, calculate a priority order of processing a task associated with a task identifier in the communication information means depending on a type of instruction executed by the task, wherein
the communication control means reads the task identifier from the communication information means on the basis on the priority order.
The scheduling system according to Supplemental Note 5, wherein
the predetermined priority order assignment method is a method that assigns a higher value of the priority order to a task with the instruction type to be the fifth instruction than a task with the instruction type to be the seventh instruction.
The scheduling system according to Supplemental Note 5, wherein
the predetermined priority order assignment method is a method that assigns a higher value of the priority order to a task that executes the fifth instruction earlier among tasks with the instruction type to be the fifth instruction.
The scheduling system according to Supplemental Note 7, wherein
the predetermined priority order assignment method is a method that assigns a higher value of the priority order to a task that executes the seventh instruction earlier among tasks with the instruction type to be the seventh instruction.
The scheduling system according to Supplemental Note 5, wherein
the predetermined priority order assignment method is a method that assigns a higher value of the priority order to a task that transmits a less amount of data via the communication channel among tasks with the instruction type to be the fifth instruction or the seventh instruction.
The scheduling system according to Supplemental Note 5, wherein
the predetermined priority order assignment method is a method that assigns a higher priority order to a task with the instruction type to be the fifth instruction than a task with the instruction type to be the seventh instruction when a storage area in an unused state capable of storing the data in the accelerator memory is decided to exist, and assigns a higher value of the priority order to a task with the instruction type to be the seventh instruction than a task with the instruction type to be the fifth instruction when the storage area is decided not to exist.
An operating system that includes the scheduling system according to any one of Supplemental Notes 1 to 10.
A scheduling method comprising:
reserving a second communication channel, that is capable of transmitting/receiving a first data processed by a task between the memory and the accelerator memory, as a second communication resource in accordance with a fifth instruction for reserving the second communication channel from a first communication channel that is capable of transmitting/receiving data between a many-core accelerator to be a resource and a processor which controls the resource; and
determining a specific resource for processing the task by referring to the first data transmitted/received via the second communication resource in accordance with a first instruction for reserving the resource; wherein
the task is processed by a calculation processing apparatus which includes the many-core accelerator, an accelerator memory accessed by the many-core accelerator, the processor, a memory accessed by the processor, and a first communication channel.
A recording medium storing a scheduling program that causes a computer to realize a scheduling function, the function comprising:
reserving a second communication channel, that is capable of transmitting/receiving a first data processed by a task between the memory and the accelerator memory, as a second communication resource in accordance with a fifth instruction for reserving the second communication channel from a first communication channel that is capable of transmitting/receiving data between a many-core accelerator to be a resource and a processor which controls the resource; and
determining a specific resource for processing the task by referring to the first data transmitted/received via the second communication resource in accordance with a first instruction for reserving the resource; wherein
the task is processed by a calculation processing apparatus which includes the many-core accelerator, an accelerator memory accessed by the many-core accelerator, the processor, a memory accessed by the processor, and a first communication channel.
The present invention has been described with the aforementioned exemplary embodiments as exemplary examples. However, the present invention is not limited to the aforementioned exemplary embodiments. In other words, various embodiments that can be understood by those skilled in the art may be applied to the present invention, within the scope thereof.
This application claims priority based on Japanese Patent Application No. 2013-109843 filed on May 24, 2013, the disclosure of which is hereby incorporated by reference thereto in its entirety.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2013-109843 | May 2013 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2014/001558 | 3/18/2014 | WO | 00 |
