JP5741544B2 - Cache control device, disk array device, array controller, and cache control method - Google Patents

Description

  The present invention relates to a cache control technique, and more particularly to a technique for performing cache control of cache management information.

  Recently, many storage devices such as magnetic disk devices and disk array devices have a cache control function from the viewpoint of improving performance. The cache control is realized by a high-speed memory device called a cache memory intervening access (reading and writing) to the storage device from the host (computer using the storage device).

  In addition to a high-speed but high-capacity cache memory (hereinafter referred to as “primary cache”), a configuration is also known in which a secondary memory includes a memory device that is slower than the primary cache but faster than the magnetic disk device. In the secondary cache control with such a configuration, the performance can be further improved by mounting a secondary cache having a larger capacity unit price and a larger capacity than the primary cache.

  In the secondary cache control used in the disk array device, for example, an area with a high reference frequency in a logical disk recognized by the host may be stored in the secondary cache. In this case, the read performance can be improved by accessing the secondary cache instead of accessing the low-speed physical disk (read cache). For example, a secondary cache may be used to temporarily store an amount of data that exceeds the writing speed of the physical disk. In this case, writing performance can be improved by writing to the physical disk asynchronously with writing from the host (write cache).

  In the secondary cache control, accesses from the host concentrate on a specific area of the secondary cache, so that the overall access performance may be deteriorated. Patent Documents 1 to 3 describe techniques for solving such problems (or similar problems).

  The disk array device described in Patent Document 1 distributes frequently used data in cache memories of a plurality of controllers, thereby distributing the load on the controllers.

  The information processing apparatus described in Patent Document 2 stores data in a storage area with high access efficiency based on access information of each storage apparatus in a system in which a plurality of hosts and storage apparatuses are connected to the same network.

  The storage system described in Patent Document 3 includes a plurality of primary caches and secondary caches, and realizes optimization of cache usage and high data reliability.

JP 2001-051809 A JP 2004-021557 A JP 2008-217527 A

  The data stored in the cache memory includes not only the data itself (actual data) accessed from the host but also management information for cache control. In the secondary cache control, when the access is concentrated on this management information, the overall access performance may be deteriorated. However, the above-mentioned Patent Documents 1 to 3 cannot cope with the problem caused by the concentrated access to such management information.

  A main object of the present invention is to provide a cache control device or the like that solves the above-described problems.

A cache control device according to the present invention that achieves the above object is as follows.
A first control means and a secondary cache means;
The first control means comprises a primary cache means that operates faster than the secondary cache means, and a first management information cache means that operates faster than the secondary cache means,
The secondary cache means stores real data and management information of the real data, and operates as a secondary cache of the primary cache means.
The first management information cache means stores the management information as a cache.

In addition, the disk array device according to the present invention that achieves the above-described object provides:
The cache control apparatus described above and a physical disk are provided.

And the array controller according to the present invention that achieves the above-mentioned object,
Connected to secondary cache means and physical disk,
A primary cache means operating faster than the secondary cache means, and a management information cache means operating faster than the secondary cache means,
The primary cache means acts as the primary cache for the physical disk,
The secondary cache means operates as the primary cache means and the secondary cache of the physical disk,
The management information cache means stores the management information of the actual data stored in the secondary cache means as a cache.

Furthermore, a cache control method according to the present invention that achieves the above object is as follows.
Store the actual data and the management information of the actual data in the secondary cache means that operates as the secondary cache of the primary cache means,
The management information is stored as a cache in the first management information cache means.

  The cache control device, disk array device, array controller, and cache control method according to the present invention alleviate performance degradation in cache control.

It is a figure explaining the structure of the disk array apparatus which concerns on the 1st Embodiment of this invention. It is a figure explaining the detailed structure of the secondary cache 131, and the data stored in a secondary cache. 6 is a diagram for explaining an example of secondary cache control when data is written to the disk array device 100. FIG. 2 is a diagram illustrating a detailed configuration of a controller 110. FIG. It is a figure explaining the cache control with respect to the management information of the secondary cache. It is a figure explaining the cache control with respect to the management information of the secondary cache 131 in 2nd Embodiment. It is a figure explaining the structure of the cache control apparatus which concerns on the 3rd Embodiment of this invention.

  Next, embodiments of the present invention will be described in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a diagram for explaining the configuration of a disk array device according to the first embodiment of the present invention. The disk array device 100 according to the first embodiment of the present invention includes controllers 110 and 120, secondary caches 131 and 132, and a physical disk unit 140. The host 200 is a computer that uses the disk array device 100. The host 200 recognizes the disk array device 100 as a storage device composed of logical disks LD1, LD2,..., LDx.

  Controllers 110 and 120 may be referred to as first and second control units.

  The controller 110 is an array controller corresponding to the logical disk LD1. The controller 120 is an array controller corresponding to the logical disk LD2. The controller 110 and the controller 120 may double as array controllers for other logical disks. The controller 110 includes a primary cache 111. In addition, the controller 120 includes a primary cache 121.

  The primary caches 111 and 121 are storage devices whose access speed is higher than that of the secondary caches 131 and 132 and the physical disk unit 140. The primary cache 111 is a primary cache corresponding to the logical disk LD1. The primary cache 121 is a primary cache corresponding to the logical disk LD2. The primary caches 111 and 121 may be DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory). In the following description, it is assumed that the primary caches 111 and 121 are DRAMs.

  The secondary cache 131 is a secondary cache corresponding to the logical disk LD1 and the primary cache 111. The secondary cache 132 is a secondary cache corresponding to the logical disk LD2 and the primary cache 121. The secondary caches 131 and 132 are implemented as logical storage areas on the storage medium 130. The secondary caches 131 and 132 may be mounted on a single storage medium 130 or may be mounted on a storage area in which a plurality of storage media 130 are combined. In the following description, as illustrated in FIG. 1, the secondary caches 131 and 132 are assumed to be mounted on a storage area in which a plurality of storage media 130 are combined. The secondary caches 131 and 132 are assumed to be divided into several groups (hereinafter also referred to as “ranks”) in which a plurality of storage media 130 are collected.

  The storage medium 130 is, for example, a large-capacity storage medium having a lower capacity unit price than the primary caches 111 and 121. Specifically, the storage medium 130 may be a flash memory drive or an SSD (Solid State Drive). In the following description, it is assumed that the storage medium 130 is an SSD.

  The physical disk unit 140 includes a storage medium 143 that is a physical disk in which data is finally stored in the disk array device 100. The physical disk unit 140 includes a disk group 141 corresponding to the logical disk LD1 and a disk group 142 corresponding to the logical disk LD2. The logical disks LD1 and LD2 are implemented as logical storage areas on physical disks (storage media 143) included in the disk groups 141 and 142, respectively. Each of the disk groups 141 and 142 includes one or more storage media 143. In the following description, it is assumed that the storage medium 143 is a magnetic disk device.

  Hereinafter, an outline of the write cache operation of the disk array device 100 will be described with reference to FIG.

  First, a case where the host 200 accesses to write a file to the logical disk LD1 will be described. First, the controller 110 corresponding to the logical disk LD1 receives a file ([1-1] in FIG. 1). Next, the controller 110 tries to write the received file to the primary cache 111 (same [1-2]). If there is no free space in the primary cache 111, the controller 110 writes the file to the secondary cache 131 (same [1-3]).

  Similarly, when the host 200 accesses to write a file to the logical disk LD2, first, the controller 120 corresponding to the logical disk LD2 accepts the file ([2-1]). Next, the controller 120 attempts to write the received file to the primary cache 121 (same [2-2]). If there is no free space in the primary cache 121, the controller 120 writes the file to the secondary cache 132 ([2-3]).

  In writing to the secondary caches 131 and 132, a hash function or the like is used to determine which area of the storage medium 130 that constitutes the secondary cache. In FIG. 1, by accessing (writing) the logical disks LD1 and LD2, the writing corresponding to these accesses is performed on the same rank (“*” in FIG. 1) of the storage medium 130.

  FIG. 2 is a diagram for explaining the detailed configuration of the secondary cache 131 and the data stored in the secondary cache. The secondary cache 131 is implemented as a logical storage area on the storage media 130-1 to 130-6. The secondary cache 131 is implemented on the storage media 130-1 and 130-2, implemented on the storage media 130-3 and 130-4, and the storage media 130-5 and 130-. 6 are divided into those implemented on top. Note that the secondary cache 131 may include more storage media 130 than these. Further, since the secondary cache 132 has the same configuration as the secondary cache 131, description thereof is omitted.

  The secondary cache 131 stores management information 131a and actual data 131b. The actual data 131b is actual data written from the host 200 or read out to the host 200. The management information 131a includes the location, size, and cache control status of actual data to be accessed. For example, the management information 131a may include the following information. (1) LDN (Logical Disk Number): Logical disk number (2) LSN (Logical Sector Number): Logical sector number (3) LEN (LENgth): Data length (4) State: Cache data state.

  The State (cache data state) includes, for example, information indicating whether the actual data 131b stored in the secondary cache is “written back” or “unwritten back” in the physical disk.

  Alternatively, a part of the actual data 131b stored in the secondary cache may be “written back” to the physical disk and a part may be “unwritten back”. That is, the State may include information indicating which part of the actual data 131b stored in the secondary cache has been written back to the physical disk and which part has not been written back to the physical disk. Specifically, for example, information on the start position and end position of an area that has not been written back to the physical disk in the entire real data 131b may be included. Further, a plurality of pieces of information on the unwritten area may exist for one piece of real data 131b stored in the secondary cache.

  In the following description, the management information includes information on which part of the actual data stored in the secondary cache has been written back to the physical disk and which part has not been written back to the physical disk. Shall.

  FIG. 3 is a diagram for explaining an example of secondary cache control when data is written to the disk array device 100. The secondary cache control process in the write cache operation of the disk array device 100 will be described below with reference to FIG. In the following description, it is assumed that the primary cache 111 could not be written because there was no free capacity.

  First, a case where the host 200 accesses the disk array device 100 to write the file F1 to the logical disk LD1 will be described. At the time of access, access I / O (Input Output) range information is issued. Specifically, the logical disk number (here, LD1) and logical sector number (also LSNxx) of the file F1, the length of data (file) (here, LENz1), and the like.

  Next, it is determined in which area the cache data of the file F1 is written to the secondary cache 131. Here, the hash function is calculated from the logical disk number (LD1) and the logical sector number (LSNxx), and it is determined that the data is written to the storage media 130-5 and 130-6 based on the value of the hash function. And

  Then, the actual data 131b-F1 of the file F1 and the management information 131a-F1 of the file (cache) are written as cache data to the determined write destination. The management information 131a-F1 includes information on the logical disk number LD1, the logical sector number LSNxx, the data length LENz1, and the cache data state State1 (“Entry 1” in FIG. 3). State 1 includes information indicating whether the cache data (actual data 131b-F1) is “written back” or “unwritten back” in the physical disk.

  Further, State1 may include information indicating that a part of the cache data (actual data 131b-F1) is “written back” and a part is “unwritten back”. Specifically, the State may include information on the start position and the end position of an area that has not been written back to the physical disk in the entire real data 131b-F1. Note that a plurality of pieces of information on the unwritten area may exist in the management information 131a-F1.

  Next, a case where the host 200 accesses the disk array device 100 to write the file F2 to the logical disk LD5 will be described. The controller 110 is assumed to be an array controller corresponding to the logical disk LD5 in addition to the logical disk LD1. At the time of access, as the access I / O range information, the write destination logical disk number (here, LD5) and logical sector number (also LSNyy), data (file) length (here, LENz2) Is issued).

  Next, as in the case of the file F1, it is determined in which area of the secondary cache 131 the cache data of the file F2 is to be written. Here, based on the value of the hash function calculated from the logical disk number (LD5) and the logical sector number (LSNyy), it is determined that data is written in the same area as the storage media 130-5 and 130-6, that is, the file F1. Shall.

  Then, the actual data 131b-F2 of the file F2 and the management information 131a-F2 of the file (cache) are written to the determined write destination. The management information 131a-F2 includes information on the logical disk number LD5, logical sector number LSNyy, data length LENz2, and cache data state State2 ("Entry 2" in FIG. 3). It is to be noted that State 2 includes information indicating whether it is “written back” or “unwritten back” in the physical disk, similar to the management information 131a-F1.

  Further, it is managed that State 2 may include information indicating that a part of the cache data (actual data 131b-F2) is “written back” and a part is “unwritten back”. It is the same as information 131a-F1.

  As described above, depending on the value of the hash function, access may be concentrated on a specific area of the secondary cache 131 (in this case, the storage media 130-5 and 130-6). Such access concentration may increase the access load on a specific area of the secondary cache 131.

  When the secondary cache 131 is written back to the physical disk, the state (State) of the cache data included in the management information 131a is updated from “unwritten back” to “written back”. Here, when the State includes information indicating that a part of the cache data (actual data 131b) is “written back” and a part is “unwritten back”, one of the actual data 131b. The State is updated each time the part is written back to the physical disk. When the amount of data written once in such partial write-back is small, a large number of write-back processes occur until the entire real data 131b is written back. That is, a large number of update processes for the management information 131a including the State occur. In such a case, the access load of the area where the management information 131a including the updated state of the secondary cache 131 is stored is increased even though the host does not make a new access to the logical disk. There is.

  FIG. 4 is a diagram for explaining the detailed configuration of the controller 110. In addition to the primary cache 111, the controller 110 includes a performance monitoring unit 112, a usage frequency monitoring unit 113, and a management information cache unit 114. The controller 120 has the same configuration as that of the controller 110, and thus description thereof is omitted.

  The management information cache unit 114 may be referred to as a first management information cache unit.

  The performance monitoring unit 112 monitors the load status for each rank constituting the secondary cache 131 and detects a rank in which the performance is degraded. Specifically, the performance monitoring unit 112 has a rank composed of storage media 130-1 and 130-2, a rank composed of storage media 130-3 and 130-4, and a rank composed of storage media 130-5 and 130-6. Monitor performance (multiplicity of access, response time, etc.). Then, when the index indicating the performance degradation of a certain rank exceeds a predetermined threshold value, it is determined that the performance of this rank is degraded. The processing of the performance monitoring unit 112 described above may be performed by periodically collecting information even when there is no access from the host 200.

  The usage frequency monitoring unit 113 counts the usage frequency of the management information stored in the secondary cache 131. Further, the use frequency monitoring unit 113 ranks the management information in descending order of use frequency for each rank.

  When accessed from the host 200 or when cache control processing such as writing back is performed, a part of the management information of the secondary cache 131 is stored in the controller 110 as a cache. The usage frequency monitoring unit 113 may extract the cached information that is frequently accessed and rank the management information based on the extraction result.

  The performance monitoring unit 112 and the usage frequency monitoring unit 113 may be realized by dedicated hardware or may be realized by a combination of a computer and software.

  Referring to FIG. 4, the usage frequency monitoring unit 113 includes a rank made up of storage media 130-1 and 130-2, a rank made up of storage media 130-3 and 130-4, and a storage medium 130-5 and 130-6. In each rank, the usage frequency of management information is tabulated. For example, in the rank composed of the storage media 130-1 and 130-2, the usage frequency of one piece of management information is tabulated. Further, in the rank composed of the storage media 130-3 and 130-4, the usage frequencies of the four pieces of management information are tabulated, and are ranked in descending order of usage frequency (3rd, 2nd, 1st, 4th from the left). ). Similarly, in the rank composed of the storage media 130-5 and 130-6, the usage frequencies of the three management information are tabulated, and are ranked in descending order of usage frequency (1st, 3rd, 2nd from the left). Has been.

  Similar to the primary cache 111, the management information cache unit 114 is a storage device that has a higher access speed than the secondary cache 131 and the physical disk unit 140. In the following description, it is assumed that the management information cache unit 114 is a DRAM. The management information cache unit 114 may be a part of the area of the primary cache 111 or may be an independent storage device.

  The management information cache unit 114 performs cache control by storing, as a cache, management information that is included in ranks with low performance and that is frequently used. Specifically, when the performance of a certain rank deteriorates, the management information cache unit 114 has a high usage frequency among the management information of the secondary cache 131 included in the rank (the rank of the usage frequency is higher). Stuff). Then, cache control of access to this management information is performed.

  FIG. 5 is a diagram for explaining cache control for the management information of the secondary cache 131. Hereinafter, the cache control process for the management information of the secondary cache 131 in the disk array device 100 will be described with reference to FIG.

  The performance monitoring unit 112 monitors the load status for each rank of the storage media 130-1 to 130-6 constituting the secondary cache 131. Specifically, for example, the response times of the storage media 130-1 to 130-6 are periodically acquired and accumulated. Then, an index indicating performance degradation is monitored. For example, it is monitored whether there is a rank whose average response time exceeds a predetermined threshold. The above processing is periodically performed in the background regardless of whether there is an access from the host 200.

  When accessed from the host 200 or when cache control processing such as writing back is performed, a part of the management information of the secondary cache 131 is copied to the controller 110. The usage frequency monitoring unit 113 may extract management information that is frequently accessed based on the copied information, and may rank the management information based on the extraction result.

  The processing in the case where an access is made to write a file from the host 200 to the disk array device 100 in a state where information update by the performance monitoring unit 112 and the usage frequency monitoring unit 113 is periodically performed is as follows. Explained.

  As a result of accepting this write access, it is assumed that writing to the secondary cache 131 has occurred due to a lack of free space in the primary cache 111 or the like. If the file from the host has been written to the primary cache 111 and no writing to the secondary cache 131 has occurred, the following processing is not performed.

  First, the rank in which the file is written is determined in the secondary cache 131 by calculating a hash function or the like. As a result, it is assumed here that the files are determined to be written to the storage media 130-5 and 130-6.

  At this point, it is assumed that the performance monitoring unit 112 has detected a performance degradation of the rank formed of the storage media 130-5 and 130-6 (shaded square in the performance monitoring unit 112 in FIG. 5). In this case, the controller 110 refers to the usage frequency of the management information stored in the ranks (storage media 130-5 and 130-6) whose performance is degraded, from the usage frequency monitoring unit 113. Then, the controller 110 identifies management information that is frequently used. Here, the management information having the highest usage frequency is extracted (“1” in a shaded square in the usage frequency monitoring unit in FIG. 5).

  The controller 110 stores the extracted management information as a cache in the management information cache unit 114 (shaded square in the management information cache unit 114). That is, the subsequent access to the extracted management information is performed not on the management information on the secondary cache 131 but on the management information cache on the management information cache unit 114. This reduces the load on the storage media 130-5 and 130-6 of the secondary cache 131.

  The cache control of the management information is not performed when the performance degradation is not detected in the rank in which the file from the host is written.

  The management information cache control is terminated when the performance monitoring unit 112 detects that the rank performance degradation has been resolved. That is, the management information cache on the management information cache unit 114 is written back to the management information on the secondary cache 131. The subsequent access to the rewritten management information is performed on the rewritten management information on the secondary cache 131 without performing cache control.

  The cache control process for the management information of the secondary cache 131 is performed in response to file writing from the host, but may be performed at other timings. For example, it may be performed in the background in a state where information is regularly updated by the performance monitoring unit 112 and the usage frequency monitoring unit 113. Specifically, when the performance monitoring unit 112 detects a performance degradation of a certain rank, the usage frequency monitoring unit 113 extracts management information having a high usage frequency on the rank where the performance is degraded. Then, the controller 110 may store the extracted management information in the management information cache unit 114 and start cache control.

  As described above, the disk array device 100 according to the first embodiment of the present invention mitigates performance degradation in secondary cache control. This is because, in the management information in the secondary cache 131 mounted on the storage medium 130 in which the performance degradation has been detected, the management information cache unit 114 performs cache control on the frequently used information. As a result, access to the management information on the storage medium 130 whose performance has deteriorated is not performed, and the performance deterioration is eliminated.

  Also, the disk array device 100 according to the first embodiment of the present invention realizes efficient cache control. This is because the management information of the secondary cache is smaller in size than the actual data stored in the secondary cache, so that the capacity of the high-speed but expensive management information cache unit 114 can be reduced. In particular, when a partial write-back process of actual data stored in the secondary cache is performed, a large number of update processes occur in the management information, so that the cache control efficiency is improved.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. In the second embodiment, the possibility of data loss at the time of failure is reduced in a state where cache control of management information is performed.

  FIG. 6 is a diagram for explaining cache control for management information of the secondary cache 131 in the second embodiment. Referring to FIG. 6, the controller 120 includes a management information cache unit 124. The other components of the disk array device 100 according to the second embodiment of the present invention are the same as those shown in FIGS. 1 and 4 described in the first embodiment, and thus the description thereof is omitted.

  The management information cache unit 124 may be called a second management information cache unit.

  In the second embodiment, when management information having a high use frequency on a rank whose performance is degraded is stored in the management information cache unit 114, the same management information is also stored in the management information cache unit 124 (see FIG. 6 is a shaded square in the management information cache unit 124). When the management information cache on the management information cache unit 114 is updated, the management information cache on the management information cache unit 124 may be updated at the same time. That is, the management information cache unit 114 and the management information cache unit 124 may perform management information duplication (mirroring).

  As described above, the disk array device 100 according to the second embodiment of the present invention has the same effect as the first embodiment, but reduces the possibility of data loss due to a failure or the like. This is because even when the management information cache unit 114 is a volatile memory such as a DRAM, the management information is mirrored on the management information cache unit 124, thereby managing the management information cache unit 114 or the controller 110 due to a failure. This is because loss of information can be prevented.

[Third Embodiment]
Next, a third embodiment of the present invention will be described.

  FIG. 7 is a diagram illustrating the configuration of the cache control device according to the third embodiment of the present invention. A cache control apparatus 1000 according to the third embodiment of the present invention includes a controller 110 and a secondary cache 131.

  The controller 110 includes a primary cache 111 that operates at a higher speed than the secondary cache 131 and a management information cache unit 114 that operates at a higher speed than the secondary cache 131.

  The controller 110 may be referred to as a first control unit. In addition, the management information cache unit 114 may be referred to as a first management information cache unit.

  The cache control apparatus 1000 according to the third embodiment of the present invention operates as a physical disk cache apparatus by being connected to a physical disk (not shown). The cache control apparatus 1000 performs cache control of actual data by storing actual data stored in the physical disk in the primary cache 111 and the secondary cache 131.

  The secondary cache 131 stores actual data and management information of the actual data as a cache. Then, it operates as a secondary cache of the primary cache unit 110.

  The management information cache unit 114 stores management information as a cache.

  As described above, the cache control device 1000 according to the third embodiment of the present invention reduces the performance degradation in the secondary cache control. This is because management information in the secondary cache 131 is cache-controlled in the management information cache unit 114. As a result, the access to the management information is performed with respect to the management information cache unit 114 operating at a higher speed, so that the performance degradation is eliminated.

100 disk array device 110, 120 controller 111, 121 primary cache 112 performance monitoring unit 113 usage frequency monitoring unit 114, 124 management information cache unit 130, 130-1 to 130-6, 143 storage medium 131, 132 secondary cache 131a, 131a-F1, 131a-F2 management information 131b, 131b-F1, 131b-F2 real data 140 physical disk unit 141, 142 disk group 1000 cache controller LD1, LD2, LD5, LDx logical disk F1, F2 file

Claims (8)

A first control means and a secondary cache means;
The first control means includes a primary cache means that operates faster than the secondary cache means, a first management information cache means that operates faster than the secondary cache means , a performance monitoring means, and a usage frequency monitoring means. including the door,
The secondary cache means stores real data and management information of the real data, and operates as a secondary cache of the primary cache means;
The first management information cache means stores the management information as a cache ;
The usage frequency monitoring means extracts management information with high usage frequency,
When the performance monitoring means monitors the load status of the secondary cache means for each rank based on the multiplicity of access to the secondary cache means, and detects the performance degradation of the secondary cache means, the first cache A cache control device, wherein the management information cache means stores, as a cache, the frequently used management information stored in the rank of the secondary cache means whose performance has deteriorated . A second control means,
The second control means includes second management information cache means;
When the first management information cache means stores the management information as a cache, the second management information cache means stores the management information, whereby the first management information cache means and the second management information cache means The cache control device according to claim 1, wherein the management information cache unit duplicates the management information. 3. The cache control according to claim 1, wherein the management information includes information indicating which part of the actual data has been written back to the physical disk and which part has not been written back to the physical disk. apparatus. The disk array device including a cache controller as claimed, a physical disk to any one of claims 1 to 3. Connected to secondary cache means and physical disk,
A primary cache means that operates faster than the secondary cache means; a management information cache means that operates faster than the secondary cache means ; a performance monitoring means and a usage frequency monitoring means ;
The primary cache means operates as a primary cache of the physical disk;
The secondary cache means operates as the primary cache means and a secondary cache of the physical disk;
The management information cache means stores management information of real data stored in the secondary cache means as a cache ,
The usage frequency monitoring means extracts management information with high usage frequency,
When the performance monitoring means monitors the load status of the secondary cache means for each rank based on the multiplicity of access to the secondary cache means, and detects a decrease in performance of the secondary cache means, the management information cache An array controller for storing, as a cache, the frequently used management information stored in the rank of the secondary cache means whose performance has deteriorated . In the secondary cache means that operates as a secondary cache of the primary cache means, real data and management information of the real data are stored,
Storing the management information as a cache in a first management information cache means ;
Extract frequently used management information,
Based on the multiplicity of access to the secondary cache means, the load status of the secondary cache means is monitored for each rank, and when the performance of the secondary cache means decreases, the first management information cache means A cache control method for storing, as a cache, the frequently used management information stored in the rank of the secondary cache means whose performance has deteriorated . When the management information is stored as a cache in the first management information cache means, the management information is stored in a second management information cache means, whereby the first management information cache means and the second management information are stored. The cache control method according to claim 6 , wherein the information cache means duplexes the management information. The cache control method according to claim 6 or 7 , wherein the management information includes information indicating which part of the actual data has been written back to the physical disk and which part has not been written back to the physical disk.

Images