JP2020140647A - Information processing method - Google Patents

Abstract

To solve the problem that a necessary data block cannot be taken out when necessary.SOLUTION: In an information processing method performed by an information processing device having a function for processing a writing request and a reading request of a file and further having a function for storing to-be-stored data, data for constituting the to-be-stored file is separated corresponding to characteristics of the data, and, according to the result of separation, the separated data is stored.SELECTED DRAWING: Figure 3

Description

The present invention relates to an information processing method, an information processing device, and a program.

A large-capacity storage system is required for the purpose of responding to the ever-increasing amount of data to be stored. In such a storage system, for example, the storage cost is reduced by eliminating duplicate storage.

As an example of the storage system as described above, there is, for example, Patent Document 1. Patent Document 1 describes a storage system having a writing unit that stores data in a deduplicated state and a data separating unit that separates data to be written into a user data portion and a management data portion. There is. According to Patent Document 1, the writing unit generates a file in which the user data portion is stored, and also stores the management data portion in the file as metadata associated with the user data portion stored in the file. Then, the writing unit stores the file as data to be written in the storage device.

Japanese Unexamined Patent Publication No. 2012-208650

In a storage system as described above, for example, as shown in FIG. 1, a pre-reading process in which a predetermined amount of data blocks are acquired from a read position in the background and cached for the purpose of ensuring the required throughput. It is carried out. As a result of this advance read processing, as shown in FIG. 1, the read request may be delinquent in the read processing queue that temporarily holds the read request, and the response time may be extended.

For example, as shown in Case 1 of FIG. 1, the read processing of a block whose read processing order is behind by the read processing queue is slower than the read of the block whose read processing order is earlier. Also, read requests outside the read-ahead range are added after the read processing queue. Therefore, as shown in Case 2 of FIG. 1, when a read request outside the read-ahead range is performed, the read processing queue is extended by the new read-ahead request, which promotes an increase in the waiting time. Further, these waiting times are further deteriorated and increased by acquiring the data block to be read from the storage device of another node.

As described above, there is a possibility that it takes time to retrieve the data block from the storage device. As a result, there is a problem that the required data block may not be retrieved when required. As a result, for example, when the video software directly reads the video file stored in the storage system and plays the video, the read I / O response time of the storage system is too long and the video playback is interrupted. ..

Therefore, an object of the present invention is to provide an information processing method, an information processing device, and a program that solve the problem that a necessary data block may not be taken out when necessary.

The information processing method, which is one embodiment of the present invention, in order to achieve such an object
It is an information processing method performed by an information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
The data that makes up the file to be stored is separated according to the characteristics of the data.
Depending on the result of separation, the separated data is stored.

Further, the information processing device according to another embodiment of the present invention is
An information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
It has a structure of having.

In addition, the program which is another form of the present invention
An information processing device that has a function to process file write requests and read requests and also has a function to store data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
It is a program to realize.

The present invention makes it possible to provide an information processing method, an information processing device, and a program that solves the problem that a necessary data block may not be taken out when necessary by being configured as described above. Become.

It is a figure which shows an example of the problem at the time of processing a read request. It is a block diagram which shows an example of the structure of the whole system including the storage system which concerns on 1st Embodiment of this invention. It is a block diagram which shows an example of the structure of the hybrid node shown in FIG. It is a figure which shows an example of the processing when data is separated and stored in a normal stream and a low delay stream. It is a figure for demonstrating an example of the process of writing a normal stream. It is a figure for demonstrating an example of the process of writing a low-delay stream. It is a figure which shows an example of the process of reading and combining a normal stream and a low delay stream which make up a file. It is a figure for demonstrating an example of the process of reading a low-delay stream. It is a figure which shows an example of the relationship between a normal read processing queue and a priority read processing queue. It is a flowchart which shows an example of the process when a storage system writes data. It is a flowchart which shows an example of the process when a storage system reads data. It is a flowchart which shows an example of the processing of a normal read processing queue and a priority read processing queue. It is a block diagram which shows an example of the structure of the information processing apparatus which concerns on 2nd Embodiment of this invention.

[First Embodiment]
The first embodiment of the present invention will be described with reference to FIGS. 2 to 12. FIG. 2 is a block diagram showing an example of the configuration of the entire system including the storage system 100. FIG. 3 is a block diagram showing an example of the configuration of the hybrid node 110. FIG. 4 is a diagram showing an example of processing when data is separated and stored in a normal stream and a low delay stream. FIG. 5 is a diagram for explaining an example of a process of writing a normal stream. FIG. 6 is a diagram for explaining an example of a process of writing a low delay stream. FIG. 7 is a diagram showing an example of a process of reading and combining a normal stream and a low-delay stream constituting a file. FIG. 8 is a diagram for explaining an example of a process of reading a low-delay stream. FIG. 9 is a diagram showing an example of the relationship between the normal read processing queue and the priority read processing queue. FIG. 10 is a flowchart showing an example of processing when the storage system 100 writes data. FIG. 11 is a flowchart showing an example of processing when the storage system 100 reads data. FIG. 12 is a flowchart showing an example of processing between the normal read processing queue and the priority read processing queue.

In the first embodiment of the present invention, the storage system 100 that stores the block data obtained by dividing the data to be stored in a deduplicated state will be described. As will be described later, the storage system 100 described in the present embodiment separates the data to be stored according to the characteristics of the data and stores the data in the normal stream and the low delay stream. Further, the storage system 100 stores the data of the normal stream and the data of the low delay stream by using different methods. That is, the storage system 100 separates the data to be stored according to the characteristics of the data and stores the data according to the characteristics of the data. Further, the storage system 100 has two types of queues for storing read requests. By properly using the two types of queues, the storage system 100 enables more appropriate reading of data.

FIG. 2 shows an example of the configuration of the entire system including the storage system 100. Referring to FIG. 2, the system includes a business system 200 and a storage system 100. As shown in FIG. 2, the storage system 100 is connected to the business system 200 so as to be able to communicate with each other by using an external NIC (Network Interface Card) 31 or the like.

Further, referring to FIG. 2, the storage system 100 has a hybrid node 110 and a storage node 120. As shown in FIG. 2, the hybrid node 110 is connected to the business system 200 so as to be able to communicate with each other by using the external NIC 31. Further, the hybrid node 110 and the storage node 120 are connected so as to be able to communicate with each other by using an internal NIC 32 or the like.

Note that FIG. 2 shows an example of the configuration of the entire system. Therefore, the configuration of the entire system is not limited to the case shown in FIG. For example, the number of hybrid nodes 110 and the number of storage nodes 120 included in the storage system 100 are not limited to those shown in FIG. The storage system 100 can have an arbitrary number of hybrid nodes 110 and an arbitrary number of storage nodes 120.

The hybrid node 110 is an information processing device having a function of processing file I / O (Input / Output) such as a file write request and a file read request, and also having a function of storing block data to be stored in the storage device 116. is there. Further, the hybrid node 110 has a cache. When the block data to be read is read from the storage device 116 or another node, the hybrid node 110 can perform a pre-reading process of acquiring a predetermined amount of data blocks from the read position and storing them in the cache.

FIG. 3 shows an example of the configuration of the hybrid node 110. Referring to FIG. 3, the hybrid node 110 includes, for example, a data separation unit 111, a separation data writing unit 112, a release unit 113, a separation data reading unit 114, a separation data association unit 115, and a storage device 116. ,have.

For example, the hybrid node 110 has an arithmetic unit such as a CPU (Central Processing Unit) and a storage device. The hybrid node 110 realizes each of the above-described processing units by, for example, executing a program stored in the storage device by the arithmetic unit.

When the data separation unit 111 receives the file write request from the business system 200, the data separation unit 111 uses the same method as that described in Japanese Patent Application Laid-Open No. 2012-208650 to obtain the pre-separation data constituting the file to be written. Separate according to the characteristics of the data. For example, as shown in FIG. 4, the data separation unit 111 separates the data before separation into data that gives priority to reading, such as requiring an immediate response at the time of reading, and data other than that. In this way, the data separation unit 111 separates the data according to the priority of reading and the like. Further, the data separation unit 111 generates data indicating the position information of the data separated into two, the data for which reading is prioritized and the other data. Then, the data separation unit 111 stores the separated / generated data in separate streams. Specifically, the data separation unit 111 stores the data in which the data before separation is separated and gives priority to reading in the low delay stream (1), and stores the data indicating the position information of the data separated into two. Store in the low latency stream (2). Further, the data separation unit 111 stores in the normal stream data other than the data for which reading is prioritized among the data separated from the data before separation.

For example, in this way, the data separation unit 111 separates the pre-separation data into two data and generates data indicating the position of the two separated data. Then, the data separation unit 111 stores each data in the normal stream and the low delay stream. As a result, the data separation unit 111 generates a file having three streams. By dividing the stream according to the characteristics of the data in this way, it is possible to optimize the data storage method and the data reading method using the stream as a clue.

The data that gives priority to reading, such as requiring an immediate response at the time of reading, is, for example, a key frame such as an I frame in the video data. In other words, the data separation unit 111 separates, for example, a frame that does not depend on another frame that can be completed with only one frame as data that gives priority to reading. The data separation unit 111 may be configured to determine a frame or the like other than those illustrated above as data for which reading is prioritized.

The separation data writing unit 112 stores the result of data separation by the data separation unit 111 in the storage system 100. The separated data writing unit 112 performs a writing process according to each stream.

For example, the separated data writing unit 112 performs a writing process on the normal stream so that all the hybrid nodes 110 and the storage node 120 are written targets, as shown in FIG. For example, the separated data writing unit 112 divides the data of the normal stream into a plurality of block data, and calculates a unique hash value representing the data content of the block data for the divided block data. Specifically, for example, the separated data writing unit 112 calculates a hash value from the data content of the block data by using a preset hash function (for example, a cryptographic hash function such as SHA-2). .. Then, the separated data writing unit 112 has already stored the block data of the hash value calculation source in the storage system 100 based on the calculated hash value and the hash information indicating the block data stored in the storage system 100. Check if it is. For example, when the calculated hash value is included in the hash information described above, the separated data writing unit 112 determines that the storage system 100 has already stored the block data having the same contents. In this case, the separated data writing unit 112 does not newly write the block data already stored in the storage system 100. On the other hand, when the calculated hash value is not included in the hash information described above, the separated data writing unit 112 determines that the block data having the same contents is not stored in the storage system 100. In this case, the separate data writing unit 112 writes the block data to either the hybrid node 110 or the storage node 120. For example, by such a process, the separated data writing unit 112 writes the block data obtained by dividing the normal stream by distributing it to the hybrid node 110 and the storage node 120. After that, the separated data writing unit 112 associates the hash value of the written block data with the information indicating the writing position of the block data, and manages it as hash information.

The separated data writing unit 112 may be configured to further divide the block data into a plurality of fragment data and distribute and write the divided fragment data to the hybrid node 110 and the storage node 120. Further, the separated data writing unit 112 may be configured to add redundant data for restoration processing to the block data or fragment data and write the data.

Further, for example, as shown in FIG. 6, the separated data writing unit 112 divides the data stored in the low delay stream (low delay stream (1) and low delay stream (2)) as a response time priority block. .. Then, the separated data writing unit 112 copies the divided response time priority block to the storage device 116 of all the nodes having the file I / O processing function (that is, all the hybrid nodes 110 of the storage system 100). In this way, the separated data writing unit 112 performs the writing process so that all the hybrid nodes 110 have the data written in the low delay stream. The separated data writing unit 112 is configured to copy the non-overlapping response time priority block to all the hybrid nodes 110 after performing the duplication determination process in the case of the low delay stream as in the case of the normal stream. It doesn't matter.

As described above, the separated data writing unit 112 performs the writing process in a different manner for each stream. In other words, the separated data writing unit 112 distributes and writes the data of the normal stream among the files to be written to the hybrid node 10 and the storage node 20 as in the case of general deduplication processing. Further, the separated data writing unit 112 stores the low-delay stream data of the files to be written in the storage device 116 and copies the data to another hybrid node 110. As a result of such write processing, the data of the normal stream is distributed and written to the hybrid node 110 and the storage node 120. On the other hand, the data of the low delay stream will be possessed by all the hybrid nodes 110.

For example, the release unit 113 deletes the data (block data, response time priority block) stored in the storage device 116 at predetermined intervals. For example, the release unit 113 deletes the data stored in the storage device 116 that does not refer to any file stored in the hybrid node 110.

Further, the release unit 113 compares a predetermined threshold value (a value may be arbitrary) with a value indicating the free capacity of the storage device 116. Then, when the value indicating the free space of the storage device 116 is smaller than the threshold value even though the above-mentioned data is deleted, the release unit 113 is predetermined among the response time priority blocks stored in the storage device 116. Delete the response time priority block that has no reference for a certain period (any period can be set). By such processing, the copy of the response time priority block of the file that has not been accessed for a certain period of time is deleted for each hybrid node 110.

The release unit 113 may be configured to determine whether or not to delete the response time priority block based on a predetermined redundancy, a value indicating free space, and a threshold value. Further, for example, when the redundancy is lower than the specified value as a result of the deletion, the release unit 113 may be configured to replace the copy of the response time priority block to be deleted with the parity data for recovery.

When the separated data reading unit 114 receives the file reading request from the business system 200, the separated data reading unit 114 identifies the data to be read based on the file name of the read request. Then, the separated data reading unit 114 specifies the storage position of the data to be read based on the stored hash information, and acquires the data to be read from each specified position.

The separated data reading unit 114 reads, for example, according to the stream to be read. For example, as shown in FIG. 7, the separated data reading unit 114 reads a normal stream with a normal read request, and reads a low delay stream or the like with a priority read request.

As described above, the data of the normal stream is distributed and stored in the hybrid node 110 and the storage node 120. Therefore, when reading the data of the normal stream, the separated data reading unit 114 reads necessary data from each of the hybrid node 110 and the storage node 120. On the other hand, as shown in FIG. 8, the data of the low delay stream is possessed by all the hybrid nodes 110. That is, in principle, the hybrid node 110 that processes the file read request has the low-delay stream data in its own storage device 116. Therefore, the separated data reading unit 114 reads the data of the low delay stream from its own storage device 116. This makes it possible to eliminate the cost of moving data between nodes when reading data from a low-latency stream.

After reading various data by the separated data reading unit 114, the separated data associating unit 115 combines various data based on the data indicating the positions of the two separated data. As a result, the separated data association unit 115 restores the file which is the original data before the separation. After that, the hybrid node 110 returns the restored file to the business system 200. As a result, the business system 200 can refer to the file without being aware that the data is separated and stored.

The separated data reading unit 114 has two types of queues for holding read requests. Referring to FIG. 9, the separated data reading unit 114 has two queues, a normal read processing queue and a priority read processing queue. For example, the separated data reading unit 114 properly uses the normal read processing queue and the priority read processing queue according to the characteristics of the block to be read and the cache state of the block. For example, when the separated data reading unit 114 reads a normal stream with a normal read request, the separated data reading unit 114 stores the read request in the normal read processing queue. On the other hand, the separated data reading unit 114 stores the read request in the priority read processing queue when the low delay stream is read by the priority read request. For example, when the separated data reading unit 114 reads a data block belonging to a low delay stream (response time priority block), or when reading data of a file system such as Btree, the data block in the uncached state of the read I / O range. The read request can be stored in the priority read processing queue when reading the data.

When a read request is stored in each of the normal read processing queue and the priority read processing queue, as shown in FIG. 9, data reading is preferentially executed from the priority read processing queue. For example, in the case shown in FIG. 9, the separated data reading unit 114 reads the blocks 1, 4, and n stored in the priority read processing queue, and then executes the processing stored in the normal read processing queue. In addition, read requests for the same block may be stored in both the normal read processing queue and the priority read processing queue. In this case, the separated data reading unit 114 deletes the other read request at the timing when the read request taken out from one queue is processed. For example, in the case of FIG. 9, the separation data reading unit 114 deletes the read request of n blocks stored in the normal read processing queue at the timing when the read request of n blocks stored in the priority read processing queue is processed. To do.

The storage device 116 is a storage device such as a hard disk. Block data, response time priority block, and the like are stored in the storage device 116.

The above is an example of the configuration of the hybrid node 110.

The storage node 120 has a storage device 121 that stores block data in a deduplicated state. The storage device 121 included in the storage node 120 can store block data, block data, parity data of a response time priority block, and the like in response to a request from the hybrid node 110. Further, the storage node 120 transmits the block data, the block data, the parity data of the response time priority block, and the like stored in the storage device 121 to the hybrid node 110 in response to the request from the hybrid node 110.

The storage node 120 can have a duplicate confirmation function for confirming whether or not the block data to be stored is stored in the storage device 121. The storage node 120 may be configured to store block data or the like to be stored in the storage device 121 after confirming duplication.

The business system 200 is an information processing device that transmits a file write request and a file read request to the hybrid node 110. In the present embodiment, the method of selecting the hybrid node 110 that executes the file write request and the file read request is not particularly limited. A file write request or read request transmitted by the business system 200 can be allocated to any of the hybrid nodes 110 of the storage system 100 by any method.

Subsequently, an example of the operation of the storage system 100 will be described with reference to FIGS. 10 to 12. First, with reference to FIG. 10, an example of the operation of the storage system 100 when a file write request is received from the business system 200 will be described.

The hybrid node 11 receives a file write request from the business system 200. Referring to FIG. 10, the data separation unit 111 separates the pre-separation data constituting the file to be written according to the characteristics of the data. For example, the data separation unit 111 separates the pre-separation data into data that gives priority to reading, such as requiring an immediate response at the time of reading, and other data. Further, the data separation unit 111 generates data indicating the position information of the data separated into two, the data for which reading is prioritized and the other data. Then, the data separation unit 111 stores the separated / generated data in separate streams (step S101). Specifically, the data separation unit 111 stores the data in which the data before separation is separated and gives priority to reading in the low delay stream (1), and stores the data indicating the position information of the data separated into two. Store in the low latency stream (2). Further, the data separation unit 111 stores in the normal stream data other than the data for which reading is prioritized among the data separated from the data before separation.

The separated data writing unit 112 divides the data stored in the low delay stream (low delay stream (1) and low delay stream (2)) as a response time priority block. Then, the separated data writing unit 112 copies the divided response time priority block to the storage device 116 of all the nodes having the file I / O processing function (that is, all the hybrid nodes 110 of the storage system 100) (that is, the storage system 100 has all the hybrid nodes 110). Step S102).

Further, the separated data writing unit 112 divides the data stored in the normal stream into block data. Then, the separated data writing unit 112 performs deduplication processing, and then distributes and stores the necessary block data in the hybrid node 110 and the storage node 120 (step S103).

For example, as described above, the separate data writing unit 112 stores data by using different methods for the low-delay stream and the normal stream. In addition, in the process of step S102 and step S103, either process may be executed first, or may be executed in parallel.

Subsequently, an example of the operation of the storage system 100 when a file read request is received from the business system 200 will be described with reference to FIG.

The hybrid node 11 receives a file read request from the business system 200. Referring to FIG. 11, when the separated data reading unit 114 receives the file reading request from the business system 200, the separated data reading unit 114 identifies the data to be read based on the file name of the read request. Then, the separated data reading unit 114 specifies the storage position of the data to be read based on the stored hash information, and acquires the data to be read from each specified position (step S201). For example, the separated data reading unit 114 reads data by properly using the normal read processing queue and the priority read processing queue according to the characteristics of the block to be read.

The separated data association unit 115 combines various data based on the data indicating the positions of the two separated data (step S202). As a result, the separated data association unit 115 restores the file which is the original data before the separation. After that, the hybrid node 110 returns the restored file to the business system 200.

Subsequently, the proper use of the normal read processing queue and the priority read processing queue will be described with reference to FIG.

Referring to FIG. 12, when the read request is stored in the priority read processing queue (step S301, Yes), the separated data reading unit 114 executes the read request stored in the priority read processing queue (step S302). .. On the other hand, when the read request is not stored in the priority read processing queue (step S301, No), the separated data reading unit 114 executes the read request stored in the normal read processing queue (step S303).

For example, as described above, the separated data reading unit 114 preferentially executes the read request stored in the priority read processing queue.

As described above, the hybrid node 110 has a data separation unit 111 and a separation data writing unit 112. With such a configuration, the separated data writing unit 112 can store data according to the result of separation by the data separating unit 111 according to the characteristics of the data. As a result, the data can be stored in the storage system 100 so that the response time is shortened. As a result, it is possible to solve the problem that the required data block may not be retrieved when required.

Further, according to the present embodiment, the separated data writing unit 112 is configured to copy the response time priority block obtained by dividing the data stored in the low delay stream to the storage devices 116 of all the hybrid nodes 110. As a result, it is guaranteed that the response time priority block to be read is stored in the storage device 116 of the hybrid node 110 that executes the file read request. As a result, it is possible to eliminate the need to transfer data between nodes when reading the response time priority block, and it is possible to reduce the time and cost required for reading. Also, copying is limited to the hybrid node 110 and not to the storage node 120. Therefore, the increase in the amount of data due to copying can be suppressed as compared with the case of copying to the entire data including the storage node 120.

Further, the hybrid node 110 has a release unit 113. With such a configuration, when the value indicating the free space of the storage device 116 is smaller than the threshold value, the release unit 113 does not refer to the response time priority block stored in the storage device 116 for a predetermined period of time. The response time priority block can be deleted or replaced with parity data. As a result, it is possible to suppress a decrease in the free space of the storage device 116 by storing the response time priority block that is unlikely to be read. That is, the storage device 116 can be used efficiently.

Further, the hybrid node 110 has a separate data reading unit 114. Further, the separated data reading unit 114 is configured to properly use the normal read processing queue and the priority read processing queue. With such a configuration, the separated data reading unit 114 can preferentially read the low-delay stream stored in the priority read processing queue. As a result, it becomes possible to preferentially read the data that needs to be read preferentially, and it is possible to solve the problem that the necessary data block may not be taken out when necessary.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIG. In the third embodiment, the outline of the configuration of the information processing apparatus 40 will be described.

The information processing device 40 has a function of processing a file write request and a file read request, and also has a function of storing data to be stored. FIG. 13 shows an example of the configuration of the information processing device 40. Referring to FIG. 13, the information processing apparatus 40 has, for example, a data separation unit 41 and a writing unit 42.

For example, the information processing device 40 has an arithmetic unit such as a CPU and a storage device. For example, the information processing device 40 realizes each of the above-mentioned processing units by executing the program stored in the storage device by the arithmetic unit.

The data separation unit 41 separates the data constituting the file to be stored according to the characteristics of the data.

The writing unit 42 stores the separated data according to the result of the separation.

As described above, the information processing device 40 has a data separation unit 41 and a writing unit 42. With such a configuration, the writing unit 42 can store the separated data according to the result of the separation by the data separating unit 41. As a result, it is possible to store the data so that the response time is shortened. As a result, it is possible to solve the problem that the required data block may not be retrieved when required.

Further, the information processing device 40 described above can be realized by incorporating a predetermined program into the information processing device 40. Specifically, the program, which is another embodiment of the present invention, has a function of processing a file write request and a file read request, and also has a function of storing data to be stored. This is a program for realizing a data separation unit 41 that separates data constituting a file to be stored according to the situation, and a writing unit 42 that stores the separated data according to the result of the separation.

Further, the information processing method executed by the above-mentioned information processing device 40 has a function of processing a file write request and a file read request, and is performed by an information processing device having a function of storing data to be stored. In this method, the data constituting the file to be stored is separated according to the characteristics of the data, and the separated data is stored according to the result of the separation.

Even the invention of the program or the information processing method having the above-mentioned configuration can achieve the above-mentioned object of the present invention because it has the same action and effect as the above-mentioned information processing apparatus 40.

<Additional notes>
Part or all of the above embodiments may also be described as in the appendix below. The outline of the information processing method and the like in the present invention will be described below. However, the present invention is not limited to the following configurations.

(Appendix 1)
It is an information processing method performed by an information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
The data that makes up the file to be stored is separated according to the characteristics of the data.
An information processing method that stores separated data according to the result of separation.
(Appendix 2)
The information processing method described in Appendix 1
An information processing method that separates the data constituting the file according to the characteristics of the data indicating whether or not to give priority to reading.
(Appendix 3)
The information processing method according to Appendix 1 or Appendix 2.
After separating the data constituting the file according to the characteristics of the data, each separated data is stored in a different stream.
An information processing method for storing data in the stream in a different way for each stream.
(Appendix 4)
The information processing method described in Appendix 3
The information processing device is connected to a storage device having a function of storing data to be stored and another information processing device.
An information processing method for copying data in a stream having the characteristics of the data to another information processing device.
(Appendix 5)
The information processing method according to any one of Supplementary note 1 to Supplementary note 4.
An information processing method that releases stored data according to the free space of the storage device.
(Appendix 6)
The information processing method according to any one of Supplementary note 1 to Supplementary note 5.
An information processing method in which a request for reading stored data is stored in one of two types of queues having different characteristics.
(Appendix 7)
The information processing method described in Appendix 6
Information that, when a process is stored in a predetermined queue of the above two types of queues, the process stored in the predetermined queue is executed with priority over the process stored in the other queue. Processing method.
(Appendix 8)
The information processing method according to any one of Supplementary note 6 or Supplementary note 7.
An information processing method for storing a request for reading the data in the two types of queues according to the characteristics of the data to be read.
(Appendix 9)
An information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
Information processing device with.
(Appendix 10)
An information processing device that has a function to process file write requests and read requests and also has a function to store data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
A program to realize.

The programs described in each of the above embodiments and appendices may be stored in a storage device or recorded in a computer-readable recording medium. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

Although the present invention has been described above with reference to each of the above embodiments, the present invention is not limited to the above-described embodiments. Various changes that can be understood by those skilled in the art can be made to the structure and details of the present invention within the scope of the present invention.

100 Storage system 110 Hybrid node 111 Data separation unit 112 Separated data writing unit 113 Release unit 114 Separated data reading unit 115 Separated data association unit 116 Storage device 120 Storage node 121 Storage device 200 Business system 31 External NIC
32 Internal NIC
40 Information processing device 41 Data separation unit 42 Writing unit

Claims (10)

It is an information processing method performed by an information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
The data that makes up the file to be stored is separated according to the characteristics of the data.
An information processing method that stores separated data according to the result of separation. The information processing method according to claim 1.
An information processing method that separates the data constituting the file according to the characteristics of the data indicating whether or not to give priority to reading. The information processing method according to claim 1 or 2.
After separating the data constituting the file according to the characteristics of the data, each separated data is stored in a different stream.
An information processing method for storing data in the stream in a different way for each stream. The information processing method according to claim 3.
The information processing device is connected to a storage device having a function of storing data to be stored and another information processing device.
An information processing method for copying the data of the stream whose characteristics of the data are predetermined to the other information processing apparatus. The information processing method according to any one of claims 1 to 4.
An information processing method that releases stored data according to the free space of the storage device. The information processing method according to any one of claims 1 to 5.
An information processing method in which a request for reading stored data is stored in one of two types of queues having different characteristics. The information processing method according to claim 6.
Information that, when processing is stored in a predetermined queue of the above two types of queues, the processing stored in the predetermined queue is executed with priority over the processing stored in the other queue. Processing method. The information processing method according to any one of claims 6 and 7.
An information processing method for storing a request for reading the data in one of the two types of queues according to the characteristics of the data to be read. An information processing device that has a function of processing a file write request and a file read request and also has a function of storing data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
Information processing device with. An information processing device that has a function to process file write requests and read requests and also has a function to store data to be stored.
A data separator that separates the data that makes up the file to be stored according to the characteristics of the data,
A writing unit that stores the separated data according to the result of the separation,
A program to realize.

Images