JP2015185105A - information processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing system which suppresses reduction in the performance of a storage device having a secondary cache.SOLUTION: An information processing system comprises: a storage device 100 which has a disk 103 for storing data, and a primary cache 101 and a secondary cache 102 for storing a part of the data stored in the disk; a data processing part 110 which processes the data stored in the disk and performs input/output of the data with respect to the primary cache; and a data control part 120 which controls the storage location of the data in the storage device in accordance with the input/output situation of the data in the data processing part. The data control part 120 determines whether or not the type of data released from the primary cache by the data processing part is registered in advance and controls whether to store the data released from the primary cache in the secondary cache in accordance with the determination result.

Description

  The present invention relates to an information processing system, and more particularly to a storage apparatus having a disk, a primary cache, and a secondary cache.

  In recent years, SSD (Solid State Drive) has become widespread as a storage medium replacing HDD (Hard Disk Drive). While SSDs have greatly improved I / O performance compared to HDDs, they are smaller and more expensive than HDDs in terms of capacity. For this reason, there are many configurations in which current storage devices use SSDs as secondary caches, and it is believed that there will be a great demand for such configurations in the future.

  Under such circumstances, the configuration of replacing all HDDs of storage devices with SSDs is disadvantageous in that it is too expensive in terms of price. In addition, there is a technology that caches memory in the application layer as a configuration that performs all processing on memory, but it has not been fully adopted because it is necessary to make the application compatible with memory cache technology and products. . The configuration in which the entire processing in the database engine is performed on the memory is also expected to be fixed due to the relationship between the memory price and the amount of data to be processed.

JP 2011-204060 A

  As described above, the SSD has excellent input / output performance, but is expensive, and thus is used as a secondary cache of the storage apparatus as described above. Here, since the SSD has a limit on the number of times of writing to a block, the writing process is limited. For example, in Patent Document 1, when discarding a block from the primary cache, it is determined whether or not the read count for the block exceeds a threshold, and if the threshold is exceeded, the block is stored in the SSD. This reduces the number of writes to the SSD and extends the life of the SSD.

  On the other hand, depending on the processing of the database, there is a situation where the effect of placing the data block in the secondary cache cannot be obtained. In such a situation, not only the effect of using the secondary cache is not effective, but also the processing in other storages is hindered and the performance is reduced.

  Therefore, an object of the present invention is to solve that the performance of a storage device having a secondary cache is degraded.

An information processing system according to one aspect of the present invention
A storage device comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk;
A data processing unit that processes data stored in the disk and inputs / outputs data to / from the primary cache;
A data control unit that controls a storage location of data in the storage device according to a data input / output situation in the data processing unit,
The data control unit determines whether or not the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in next cache,
The configuration is as follows.

The program which is the other form of this invention is:
A storage device comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk;
A data processing unit that processes data stored in the disk and inputs / outputs data to / from the primary cache;
To the information processing device connected to
According to the data input / output status in the data processing unit, a program for realizing a data control unit that controls a data storage location in the storage device,
The data control unit determines whether or not the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in next cache,
The configuration is as follows.

A cache control method according to another aspect of the present invention includes:
A storage device cache control method comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk,
The data processing unit processes the data stored in the disk and inputs / outputs data to / from the primary cache.
According to the data input / output status of the data processing unit, the storage location of the data in the storage device is controlled, and whether the type of data released from the primary cache by the data processing unit is registered in advance. Control whether to store the data released from the primary cache in the secondary cache according to the determination result,
The configuration is as follows.

  By configuring as described above, the present invention can suppress a decrease in performance of a storage apparatus having a secondary cache.

It is a block diagram which shows the structure of the information processing system in this invention. It is a figure for demonstrating operation | movement of the information processing system disclosed in FIG. It is a figure for demonstrating operation | movement of the information processing system disclosed in FIG. It is a figure for demonstrating operation | movement of the information processing system disclosed in FIG. It is a figure for demonstrating operation | movement of the information processing system disclosed in FIG. It is a figure for demonstrating operation | movement of the information processing system disclosed in FIG. It is a block diagram which shows the structure of the information processing system in attachment 1 of this invention.

  A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing the configuration of the information processing system, and FIGS. 2 to 6 are diagrams for explaining the operation.

  As shown in FIG. 1, the information processing system includes a storage device 10, a database program 20, and an operation management program 30. This information processing system may be configured by, for example, one information processing apparatus including an arithmetic device, or may be configured by a plurality of information processing apparatuses.

  As illustrated in FIG. 1, the storage device 10 includes a logical disk 13 that is a storage device configured by an HDD (Hard Disk Drive). The storage apparatus 10 also includes a primary cache 11 and a secondary cache 12 that store a part of data stored in the logical disk 13. The secondary cache 12 is configured by, for example, an SSD (Solid State Drive), and the read / write speed is faster than the logical disk 13 but slower than the primary cache 11.

  The database program 20 is a data processing unit realized by reading and executing the program on an arithmetic device provided in the information processing apparatus. The database program 20 stores data stored in the logical disk 13. Process and input / output data to / from the primary cache. Specifically, as shown in FIG. 1, the database program 20 includes a process group that performs various processes such as a database process 21, and memory areas such as a database buffer 22, a log buffer 23, and a database instance management area 24. Yes.

  The operation management program 30 is a data control unit realized by reading and executing the program in an arithmetic unit provided in the information processing apparatus. The operation management program 30 stores data according to data input / output status in the database program 20. The storage location of data in the apparatus 10 is controlled. In particular, in this embodiment, the operation management program 30 controls whether or not data released from the primary cache by the database program 20 is stored in the secondary cache. Hereinafter, each configuration and its operation will be described in detail.

  First, the database program 20 processes the data stored in the logical disk 13 via the primary cache 11 as shown by the arrows in FIG. 2, and inputs / outputs data to / from the primary cache 11 at that time. The data released from the primary cache 11 by the data program 20 is written back to the logical disk 13 by the operation management program 30 or discarded as it is, as will be described later, or the secondary cache 12 The memory location is controlled so that it is written in the memory. Hereinafter, a method for controlling data released from the primary cache 11 by the data program 20 will be described in detail.

  The database program 20 always stores the data usage status by the database process 21 in the database instance management area 24. The database process 21 is a process such as a process for writing out the data buffer 22, a process for writing out the transaction log 23, or a process for processing a request from the user.

  As shown in FIG. 4, the database instance management area 24 includes DB object placement information 24a, process input / output statistics 24b, session information 24c, and buffer management information 24d as data utilization status by the database program 20. Remembered. The DB object arrangement information 24a is an object name including the type of data to be input / output by the database program 20, and disk position information that is the input / output location thereof. The process input / output statistics 24cb is disk position information that is an input / output location of data to be input / output by the database program 20, and an input / output size that represents the size of the input / output data. The session information 24c is information representing connection (session) identification information (ID) by a user who issues a processing request for data, and the status of whether or not the connection (session) is still present. The buffer management information 24d includes an object name including the type of data stored in the buffer, information indicating whether in-memory processing is applied to the object, a keep attribute that is an attribute that is forcibly held in the buffer, and Information indicating whether or not.

  The use status of data stored in the database instance management area 24 of the data program 20 is read by the database monitoring unit 32 of the operation management program 30 (arrow S1 in FIG. 3, arrow S11 in FIG. 4, arrow in FIG. 5). S22). Then, based on the use status of the read data, the determination unit 33 determines whether or not to use the secondary cache 12 as a data storage destination of the primary cache 11 released by the database program 20. (Arrow S2 in FIG. 3). At this time, the determination unit 33 determines whether or not to use the secondary cache 12 based on the determination criterion registered in the registration information area 31 in advance and the usage status of the data stored in the database instance storage area 24. (Arrows S23 and S24 in FIGS. 5 and 6).

  Here, in the registration information area 31, for example, the information shown in FIG. 6 is registered in advance by the user as a criterion (arrow S21 in FIG. 5). As shown in FIG. 6, the determination criterion indicates “input / output target” including the type of data to be input / output to / from the database program, and “whether the secondary cache 12 is used for the input / output target”. “Use decision” and “judgment logic” indicating the situation of whether or not to apply the decision.

  Specifically, the determination unit 33 determines the storage destination of the data released from the primary cache 11 as follows from the usage status of the data read by the database monitoring unit 32 using the determination criteria shown in FIG. To do. First, when the object name of the data read by the database monitoring unit 32 is “transaction log” or “archive log”, it is determined that the data is not reused, and “secondary cache 12 is not used. ". If the object name of the data read by the database monitoring unit 32 is “table data” or “index”, and the usage status is “keep in buffer”, the data is stored in the cache Therefore, it is determined that “the secondary cache 12 is not used”. On the other hand, if the data usage status is “not kept in the buffer”, it is determined that “use the secondary cache 12” in preparation for the case where the data is reused.

  If the object name of the data read by the database monitoring unit 32 is “table data” or “index” and the usage status is “in-memory applied”, the data is stored in the cache. Therefore, it is determined that “the secondary cache 12 is not used”.

  Also, when the object name of the data read by the database monitoring unit 32 is “table data” or “index”, and the usage status is “table data creation (load)” or “index creation / reproducibility” Since it is not necessary to store data in the cache, it is determined that “the secondary cache 12 is not used”. At this time, in particular, when the data is the above-described object name and the size of the read data is equal to or larger than a predetermined value, it is determined that “the secondary cache 12 is not used”, and otherwise “2” It is determined that the next cache 12 is used.

  Further, when the object name of the data read by the database monitoring unit 32 is “temporary table data” and the usage status is “session persists”, the data is stored in the cache. Since it is not necessary, it is determined that “the secondary cache 12 is not used”. On the other hand, when the data usage status is “session has ended”, it is determined that “use secondary cache 12” in preparation for the case where data is reused.

  In addition, when the object name of the data read by the database monitoring unit 32 is not registered as a determination criterion, or when the usage status does not satisfy the registered determination criterion even if the object name is registered, As a default determination result, it is determined that “use the secondary cache 12”.

  As described above, in principle, the determination unit 33 determines to store data released from the primary cache 11 in the secondary cache 12 when there is a possibility of being reused. On the other hand, the determination unit 33 is less likely to be reused when the type of data released from the primary cache 11 is registered and the usage status is registered. It is determined that storage in the cache 12 is unnecessary, and determination is made not to store in the secondary cache 12.

  Then, the determination unit 33 transmits the determination result as described above to the execution unit 34 (arrow S3 in FIG. 3 and arrow S25 in FIGS. 5 and 6). Then, the execution unit 34 controls the storage destination of the data released from the primary cache 11 via the storage control unit 14 of the storage apparatus 10 (arrow S4 in FIG. 3 and arrow S26 in FIG. 5). That is, as described above, the execution unit 34 performs control so that the data released from the primary cache 11 is stored in the secondary cache 12 when it is determined that “the secondary cache 12 is used”. On the other hand, when it is determined that “the secondary cache 12 is not used”, the data that has been released from the primary cache 11 is not stored in the secondary cache 12 but is written back to the logical disk 13 or discarded. To do.

  As described above, the information processing system according to the present invention controls whether or not data is stored in the secondary cache according to the type of data released from the primary cache 11 and the usage status. As a result, when data reuse is expected, data released from the primary cache 11 is stored in the secondary cache. On the other hand, if data reuse is not expected, the data released from the primary cache is discarded or written back to the disk without being stored in the secondary cache. As a result, it is possible to suppress the use of the secondary cache in an unnecessary manner, improve the utilization efficiency, and improve the performance of the entire system.

  In the information processing system described above, the storage control unit 14 provided in the storage apparatus 10 may receive and store the determination criterion stored in the operation management program 30. Then, when the data is discarded from the primary cache 11, the storage control unit 14 may determine whether or not to use the secondary cache 12 according to the data usage status and the stored determination criteria. .

  Here, the database described above may be a database having any structure, and may be a product of a specific company or an open source product.

<Appendix>
Part or all of the above-described embodiment can be described as in the following supplementary notes. The outline of the configuration of the information processing system (see FIG. 7), program, and cache control method according to the present invention will be described below. However, the present invention is not limited to the following configuration.

(Appendix 1)
A storage apparatus 100 comprising a disk 103 for storing data, and a primary cache 101 and a secondary cache 102 for storing a part of the data stored in the disk;
A data processing unit 110 that processes data stored in the disk and inputs / outputs data to / from the primary cache;
A data control unit 120 that controls a storage location of data in the storage device in accordance with data input / output status in the data processing unit,
The data control unit 120 determines whether the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in secondary cache,
Information processing system 100.

(Appendix 2)
An information processing system according to attachment 1, wherein
The data control unit does not store the type of data released from the primary cache in the secondary cache when it is determined that it is registered, and when it is determined that it is not registered, Control to store in secondary cache,
Information processing system.

(Appendix 3)
An information processing system according to appendix 1 or 2,
Whether the data control unit is a status in which the usage status by the data processing unit is registered in advance among the data determined that the type of data released from the primary cache is registered And whether to store the data released from the primary cache in the secondary cache according to the determination result,
Information processing system.

(Appendix 4)
An information processing system according to attachment 3, wherein
The data control unit is released from the primary cache when the usage status of the data released from the primary cache is a registered status that is held in a buffer provided in the data processing unit. Store the data released from the primary cache in the secondary cache, and store the data released from the primary cache in the secondary cache.
Information processing system.

(Appendix 5)
An information processing apparatus according to appendix 3 or 4, wherein
The data control unit is released from the primary cache when the usage status of the data released from the primary cache is a registered status in which in-memory processing is applied in the data processing unit. Store the data released from the primary cache in the secondary cache, and store the data released from the primary cache in the secondary cache.
Information processing system.

(Appendix 6)
An information processing apparatus according to any one of appendices 3 to 5,
The data control unit, when the usage status of the data released from the primary cache is a registered status that a session exists in the data processing unit, the data released from the primary cache Is not stored in the secondary cache, otherwise the data released from the primary cache is stored in the secondary cache.
Information processing system.

(Appendix 7)
An information processing apparatus according to any one of appendices 3 to 6,
The data control unit, when the usage status of the data released from the primary cache is a registered status in which a data size of a predetermined value or more is read by the data processing unit, from the primary cache Not storing the released data in the secondary cache, otherwise storing the released data in the secondary cache in the secondary cache;
Information processing system.

  According to the above invention, first, the data processing unit inputs data read from the disk and stored in the primary cache, and processes the data. When the data that is no longer needed by the data processing unit is released from the primary cache, the data control unit determines whether the type of the data is registered in advance. Then, the data control unit controls whether to store in the secondary cache according to the determination result of the data released from the primary cache. For example, data whose type is registered is judged not to be reused, and is written back to the disk or discarded without being stored in the secondary cache. On the other hand, data whose type is not registered is determined to be reused and stored in the secondary cache. In addition to the determination result of whether or not the type of data is registered, the data control unit determines whether or not the data usage status by the data processing unit is registered, and reuse is performed according to the determination result. It may be determined whether or not the data is expected and whether or not the data released from the primary cache is stored in the secondary cache may be controlled. As a result, it is possible to suppress the use of the secondary cache in vain, improve the utilization efficiency, and improve the performance of the entire system.

(Appendix 8)
A storage device comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk;
A data processing unit that processes data stored in the disk and inputs / outputs data to / from the primary cache;
To the information processing device connected to
According to the data input / output status in the data processing unit, a program for realizing a data control unit that controls a data storage location in the storage device,
The data control unit determines whether or not the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in next cache,
program.

(Appendix 9)
A storage device cache control method comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk,
The data processing unit processes the data stored in the disk and inputs / outputs data to / from the primary cache.
According to the data input / output status of the data processing unit, the storage location of the data in the storage device is controlled, and whether the type of data released from the primary cache by the data processing unit is registered in advance. Control whether to store the data released from the primary cache in the secondary cache according to the determination result,
Cache control method.

(Appendix 10)
The cache control method according to attachment 9, wherein
Of the data determined to be registered as the type of data released from the primary cache, it is further determined whether or not the usage status by the data processing unit is registered in advance, and the determination result In response to controlling whether to release the data released from the primary cache to the secondary cache,
Cache control method.

  Note that the above-described program is stored in a storage device or recorded on 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 with reference to the above-described embodiment and the like, the present invention is not limited to the above-described embodiment. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

DESCRIPTION OF SYMBOLS 10 Storage apparatus 11 Primary cache 12 Secondary cache 13 Logical disk 14 Storage control part 20 Database program 21 Database process 22 Database buffer 23 Log buffer 24 Database instance management area 24a DB object arrangement information 24b Process input / output statistics 24c Session information 24d Buffer management Information 30 Operation management program 31 Registration information area 32 Database monitoring unit 33 Determination unit 34 Execution unit 100 Storage device 101 Primary cache 102 Secondary cache 103 Disk 110 Data processing unit 120 Data control unit

Claims (10)

A storage device comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk;
A data processing unit that processes data stored in the disk and inputs / outputs data to / from the primary cache;
A data control unit that controls a storage location of data in the storage device according to a data input / output situation in the data processing unit,
The data control unit determines whether or not the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in next cache,
Information processing system. The information processing system according to claim 1,
The data control unit does not store the type of data released from the primary cache in the secondary cache when it is determined that it is registered, and when it is determined that it is not registered, Control to store in secondary cache,
Information processing system. The information processing system according to claim 1 or 2,
Whether the data control unit is a status in which the usage status by the data processing unit is registered in advance among the data determined that the type of data released from the primary cache is registered And whether to store the data released from the primary cache in the secondary cache according to the determination result,
Information processing system. The information processing system according to claim 3,
The data control unit is released from the primary cache when the usage status of the data released from the primary cache is a registered status that is held in a buffer provided in the data processing unit. Store the data released from the primary cache in the secondary cache, and store the data released from the primary cache in the secondary cache.
Information processing system. An information processing apparatus according to claim 3 or 4,
The data control unit is released from the primary cache when the usage status of the data released from the primary cache is a registered status in which in-memory processing is applied in the data processing unit. Store the data released from the primary cache in the secondary cache, and store the data released from the primary cache in the secondary cache.
Information processing system. An information processing apparatus according to any one of claims 3 to 5,
The data control unit, when the usage status of the data released from the primary cache is a registered status that a session exists in the data processing unit, the data released from the primary cache Is not stored in the secondary cache, otherwise the data released from the primary cache is stored in the secondary cache.
Information processing system. An information processing apparatus according to any one of claims 3 to 6,
The data control unit, when the usage status of the data released from the primary cache is a registered status in which a data size of a predetermined value or more is read by the data processing unit, from the primary cache Not storing the released data in the secondary cache, otherwise storing the released data in the secondary cache in the secondary cache;
Information processing system. A storage device comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk;
A data processing unit that processes data stored in the disk and inputs / outputs data to / from the primary cache;
To the information processing device connected to
According to the data input / output status in the data processing unit, a program for realizing a data control unit that controls a data storage location in the storage device,
The data control unit determines whether or not the type of data released from the primary cache by the data processing unit is registered in advance, and determines the data released from the primary cache according to the determination result. Control whether to store in next cache,
program. A storage device cache control method comprising: a disk for storing data; and a primary cache and a secondary cache for storing a part of the data stored in the disk,
The data processing unit processes the data stored in the disk and inputs / outputs data to / from the primary cache.
According to the data input / output status of the data processing unit, the storage location of the data in the storage device is controlled, and whether the type of data released from the primary cache by the data processing unit is registered in advance. Control whether to store the data released from the primary cache in the secondary cache according to the determination result,
Cache control method. The cache control method according to claim 9, comprising:
Of the data determined to be registered as the type of data released from the primary cache, it is further determined whether or not the usage status by the data processing unit is registered in advance, and the determination result In response to controlling whether to release the data released from the primary cache to the secondary cache,
Cache control method.

Images