KR102580545B1 - Apparatus and method for managing a disk cache - Google Patents

Abstract

The present invention provides a disk cache operating device and method, and the disk cache operating device according to the present invention includes a metadata storage unit that stores information about data stored in the disk cache; an operation monitoring unit that monitors information transmitted from the meta data storage unit; an operation log storage unit that stores an operation log monitored by the operation monitoring unit; an operation analysis unit that analyzes the operation log stored in the operation log storage unit and analyzes a utilization rate of data stored in the disk cache; and an option setting unit that sets options for data to be stored in the disk cache according to the control of the operation analysis unit and controls the disk cache according to the options.

Description

Disk cache operating device and method {APPARATUS AND METHOD FOR MANAGING A DISK CACHE}

The present invention relates to an apparatus and method for operating a disk cache provided in a storage system.

There is a big difference in the speed of processing data movement between the CPU of the device where the user application runs and the disk provided in the storage system.

A disk cache is a memory device used to increase the speed of reading data from a disk. In particular, it performs the function of preserving the contents of blocks read from storage.

When using a disk cache, data read from the disk is stored in the disk cache and then read, thereby preventing access to the disk.

When there is a lot of access to data stored on the disk, using a disk cache allows the device running the application to access the data more effectively than accessing the disk directly to access the necessary data.

However, if the data stored in the disk cache is not the data desired by the application terminal, the effectiveness of the disk cache may be reduced.

Accordingly, the purpose of the present invention is to provide a disk cache operation device and method that can change the data to be stored in the disk cache by analyzing the utilization rate at which data stored in the disk cache is actually used.

A disk cache operation device according to the present invention for achieving the above-described object includes a metadata storage unit that stores information about data stored in the disk cache; an operation monitoring unit that monitors information transmitted from the meta data storage unit; an operation log storage unit that stores an operation log monitored by the operation monitoring unit; an operation analysis unit that analyzes the operation log stored in the operation log storage unit and analyzes a utilization rate at which data stored in the disk cache is actually used; and an option setting unit that sets options for data to be stored in the disk cache according to the control of the operation analysis unit and controls the disk cache according to the options.

According to the present invention, data stored on a disk is temporarily stored in a disk cache, so that an application device can quickly use the data. To this end, the present invention can dynamically change the settings of the disk cache necessary to load frequently accessed data into the disk cache.

1 is an exemplary diagram showing a communication system to which a disk cache operation device according to the present invention is applied.
FIG. 2 is an exemplary diagram illustrating in detail the storage system and disk cache operation device shown in FIG. 1.
Figure 3 is a flowchart of an embodiment of a disk cache operation method according to the present invention.
Figures 4 and 5 are flowcharts showing an embodiment of a method for selecting an option to operate a disk cache among the disk cache operating methods according to the present invention.
Figure 6 is an example diagram showing a disk cache storage method applied to the disk cache operation method according to the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to provide common knowledge in the technical field to which the present invention pertains. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

In describing embodiments of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description will be omitted. The terms described below are terms defined in consideration of functions in embodiments of the present invention, and may vary depending on the intention or custom of the user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Hereinafter, a disk cache operation device and method according to an embodiment of the present invention will be described in detail with reference to the attached drawings.

FIG. 1 is an exemplary diagram showing a communication system to which a disk cache operating device according to the present invention is applied, and FIG. 2 is an exemplary diagram specifically showing the storage system and disk cache operating device shown in FIG. 1.

In FIG. 2, data read from the disk 220 is present in the disk cache 210, and in order to store frequently accessed data in the disk cache 210, the disk cache operating device 100 according to the present invention is configured as the disk cache 210. The settings for operating the disk cache can be changed dynamically. In particular, when reads are repeatedly made from the disk cache 210, the disk cache 210 has an important meaning. Accordingly, hereinafter, the present invention will be described taking the read operation of the disk cache 210 as an example.

A communication system to which the disk cache operating device 100 according to the present invention is applied, as shown in FIGS. 1 and 2, includes a storage system 200 including a disk cache 210 and a disk 220, and the storage An application device 300 that stores various data in the system 200 or uses data stored in the storage system 200 and analyzes the utilization rate at which data stored in the disk cache 210 is actually used, and the disk cache 210 ) and a disk cache operating device 100 that can change data to be stored.

When the application device 300 requests access to data stored on the disk 220, a disk operation that takes a long input/output time may be performed in the disk cache 210. Through this, the delay time that occurs when the application device 300 directly accesses the disk 220 can be reduced.

As the application device 300 mainly performs analysis tasks, when there is a data processing pattern in which the same data is repeatedly used, the efficiency of the disk cache 210 can be increased.

The method of accessing large amounts of data that is sequentially input/output is different from the method of accessing data depending on the reading pattern.

When the same data is used repeatedly, if the repeated data is stored in the disk cache 210, access to the data can proceed quickly during the analysis process, and thus the analysis process can proceed quickly.

However, if the data storage method of the disk cache 210, once set, does not change dynamically according to the environment of the storage system 200, the efficiency of using the disk cache 210 may be reduced.

Accordingly, the present invention can change the data storage method of the disk cache 210 by determining which information should be read and loaded into the disk cache 210.

Since data that is relatively smaller in size than the disk 220 is stored in the disk cache 210, the data read hit rate may be lower than expected. To prevent this, the present invention can redefine and reflect the data storage method of the disk cache 210.

Accordingly, data preserved in the disk cache 210 can be efficiently read by the application device 300, inappropriate data replacement can be prevented, and data access response time can be reduced. .

That is, the present invention provides a disk cache operating device and method that can set operating options between the disk 220 and the disk cache 210 and reset the operating options depending on the situation.

To explain further, the storage system 200 applied to a high-performance analysis computing environment or the storage system 200 providing a large amount of data can quickly provide data to the application device 300. ), the disk cache 210 is provided.

In this case, the disk cache operating device 100 according to the present invention provides a method of operating the disk cache 210 so that data frequently accessed in the disk cache 210 can be stored in advance in the disk cache 210. can be changed dynamically.

The disk 220 provided in the storage system 200 functions to store a large amount of data. The disk 220 mounts device files so that a storage device such as an HDD disk can access the files to find necessary data, and reads or writes data through a block I/O interface. The disk 220 can provide high capacity, but because the delay time for reading data is long, the speed at which the main memory device or CPU reads and processes data may be slow.

When data with frequently accessed characteristics is stored in the disk cache 210, the application device 300 can access the necessary data more quickly than when reading and writing by directly accessing the disk 220.

Hereinafter, the disk cache operation device 100 according to the present invention will be described in detail with reference to FIG. 2.

As shown in FIG. 2, the disk cache operation device 100 according to the present invention includes a metadata storage unit 115 that stores information about data stored in the disk cache 210, and the metadata storage unit 115. An operation monitoring unit 111 that monitors information transmitted from the operation monitoring unit 111, an operation log storage unit 112 that stores the operation log monitored by the operation monitoring unit 111, and an operation log stored in the operation log storage unit 112. an operation analysis unit 113 that analyzes the utilization rate at which the data stored in the disk cache 210 is actually used, and the data to be stored in the disk cache 210 under the control of the operation analysis unit 113. It includes an option setting unit 114 that sets options and controls the disk cache 210 according to the options.

First, the metadata storage unit 115 stores status information on whether the consistency of data stored in the disk cache 210 is maintained and data read error setting information.

Second, the operation monitoring unit 111 monitors read/write operations performed through the disk cache 210 and stores information about data read from the disk 220 to the disk cache 210. , the content stored in the disk cache stores operation operation information written to the disk 220.

Additionally, the operation monitoring unit 111 may monitor various information about the disk cache 210 using information transmitted from the meta data storage unit 115.

In addition to the operation log information of the disk cache 210, patterns of accessing data stored on or read from the disk 220 affect the operation of the disk cache 210. Accordingly, data size and data storage address information regarding I/O for block data can also be collected by the operation monitoring unit 111.

In particular, data related to read during the operation of the disk cache 210 may include disk cache read data size, data storage address, cache slot number, etc.

Third, the operation log storage unit 112 stores data monitored by the operation monitoring unit 111.

Fourth, the operation analysis unit 113 determines in real time whether the disk cache 210 is currently being used efficiently, based on operation data related to the operation of the disk cache stored in the operation log storage unit 112. Analyze.

That is, the operation analysis unit 113 determines the situation in which the disk cache 210 is being utilized based on the operation record analysis information stored in the operation log storage unit 112.

Analysis indicators used by the operation analysis unit 113 may include read hit rate, write hit rate, time stored in the cache, free space size, etc.

Fifth, the option setting unit 114 dynamically resets the options necessary for the disk cache 210 to operate.

The option setting unit 114 sets the disk cache setting value so that the read hit rate of the disk cache 210 can be improved based on the analysis information from the operation analysis unit 113, and the set value is set to the above. Applies to the disk cache (210).

Changed options are dynamically reflected in the disk cache 210. Accordingly, the disk cache 210 performs read and write operations according to the changed options.

The object of the option setting may include data pre-reading, window size, etc.

The option setting unit 114 can determine whether to continue or stop the monitored and changed settings during a critical designated time period during which the expected hit rate can be achieved.

Figure 3 is a flowchart of an embodiment of a disk cache operating method according to the present invention, and in particular, shows a method by which the disk cache operating device 100 according to the present invention changes options of the disk cache 210.

That is, the disk cache operating device 100 stores the operation log of the disk cache 210 (S100), analyzes the operation pattern of the disk cache, and selects an option to operate the disk cache 210 ( S200), options for storing data in the disk cache 210 can be reset (S300).

To elaborate, the disk cache operating device 100 monitors the operating status of the disk cache 210 (S100) and, if necessary, selects a new option to operate the disk cache 210 (S200). , the disk cache 210 can be operated using the selected option.

Figures 4 and 5 are flowcharts showing an embodiment of a method for selecting an option to operate a disk cache among the disk cache operating methods according to the present invention. That is, FIGS. 4 and 5 are exemplary diagrams showing in detail the disk cache operation pattern analysis and disk cache operation option selection process (S200) shown in FIG. 3.

Figure 6 is an exemplary diagram showing a disk cache storage method applied to the disk cache operation method according to the present invention. In particular, Figure 6 (a) shows the window size of the disk cache when the disk cache performs sequential read. This indicates an increase, and Figure 6(b) shows the read pattern of the disk cache when the disk cache performs random read.

In particular, FIGS. 4 and 5 show analysis to be applied when the application device 300 reads data stored in the disk cache 210 or when the disk cache has data to be written in the storage system 200. Indicates conditions and option setting rules.

To elaborate, FIGS. 4 and 5 show a method by which the disk cache operation device 100 analyzes and sets the operation of the disk cache 210 when the disk cache 210 performs a read operation. .

Additionally, (a) of FIG. 6 shows that when it is determined that the disk cache 210 performs sequential reads, the read window of the disk cache 210 is expanded by b1 and b2. The disk cache 210 can load and store data as much as the expanded b1 and b2 from the disk 220.

In addition, (b) of FIG. 6 shows that the disk cache 210 operates according to a random read pattern when it is determined that the disk cache 210 performs random read. The disk cache 210 may store data loaded from the disk 220 in b0, b2, b4, etc. according to a random read pattern.

First, the disk cache operating device 100 checks the address and read pattern of data read from the disk 220 to the disk cache 210 in order to check the data read pattern of the disk cache 210. (S202).

Next, sequential read is a read operation that has a high possibility of manipulating multiple blocks. Therefore, when data is read sequentially, that is, if the sequential read check result (S204) is true, the disk cache operating device 100 sets the option to pre-set the data to be read from the disk 220. The [Data Read Ahead] value of the disk cache of unit 114 is set to enabled (S208).

In addition, as shown in (a) of FIG. 6, the disk cache operating device 100 increases [window size] by α (S210) to transfer data from the disk 220 to the disk cache 210. Read data.

If the sequential read check result (S204) is false, that is, if it is determined that the disk cache 210 is performing random read, the disk cache operating device 100 determines that the disk cache 210 is performing random reads as shown in (b) of FIG. 6. As shown, the operation option of the disk cache 210 can be selected to operate according to a random pattern (S206).

To explain further, if the random read check result is true, the disk cache operation device 100 sets the [Read Data Ahead] value to enabled, and as shown in (b) of FIG. 6, random Data is read into the disk cache 210 according to the read pattern.

Next, after adjusting [window size] (S212), if the read hit rate increases (S214), the window size is maintained (S216).

However, after adjusting the [window size] (S212), if the read hit rate does not increase (S214) and the read miss rate increases (218), the disk cache operation device 100 adjusts the [window size] You can return it to its original state (S220) and set the [Read Data Ahead] value to disabled (S222).

Next, when [Window Size] maintains the initial value, that is, when the window size is not expanded (S212), the read failure rate increases (S218) when the disk cache 210 is transferred from the disk 220 to the disk cache 210. The delay time for fetching data increases, meaning that data is not ready in the disk cache 210.

Therefore, in this case, the disk cache operating device 100 reduces the [window size] to 1/2 the size (S220) so that the data reading time can be reduced, so that the divided data can be stored in the disk cache 210. You can change the settings so that it can be read (S222).

Next, when [Window size] is enabled (S210), if the read hit rate is increasing, cache operation has been improved, so [Window size] is maintained at the expanded size (S216).

Lastly, when the [Read Data Ahead] value is enabled (S208) and the read failure rate is increasing, there is no effect of reading data as much as [Window Size] to read larger data.

Accordingly, the disk cache operation device 100 reduces [window size] (S220) and initializes [data read ahead] (S222).

Additionally, the disk cache operation device 100 checks whether the free space of the disk cache is maintained (S224) and ensures that the free space maintenance rate is maintained. The disk cache operating device 100 may select data to be deleted to maintain free space and delete the selected data (S226).

As described above, the present invention stores the history generated during a read caching operation, periodically analyzes the history, and changes the settings of the disk cache 210 for caching. To this end, the disk cache operating device 100 may change the settings of the disk cache 210 according to the read pattern and read history of data.

To elaborate, the present invention can adjust the window size of the disk cache 210 according to the input/output pattern of data cached according to a read operation and the read hit rate of the cached data, and dynamically adjust the window size of the disk cache 210 ) caching options can be reset. Additionally, the present invention can optimize the disk cache 210 by pre-selecting data to be read from a block disk.

Meanwhile, although the disk cache operation device and method according to the present invention have been described according to embodiments, the scope of the present invention is not limited to the specific embodiments, and is within the scope of the present invention and is obvious to those skilled in the art. It can be implemented with various alternatives, modifications and changes.

Accordingly, the embodiments described in the present invention and the accompanying drawings are not intended to limit the technical idea of the present invention, but are for illustrative purposes, and the scope of the technical idea of the present invention is not limited by these embodiments and the attached drawings. . The scope of protection of the present invention should be interpreted in accordance with the claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of rights of the present invention.

100: Disk cache operating device
111: Operation monitoring department
112: Operation log storage unit
113: Operation analysis department
114: Option setting section
115: metadata storage unit
200: storage system
210: disk cache
220: disk
300: application device

Claims (10)

A device that operates the disk cache in a storage system including a disk cache and a disk,
a metadata storage unit that stores status information about whether the consistency of data stored in the disk cache is maintained and data read error setting information;
an operation monitoring unit that monitors an operation log for the disk cache using information transmitted from the metadata storage unit;
an operation log storage unit that stores monitoring operation log data from the operation monitoring unit;
An operation analysis unit that analyzes the operation log monitored by the operation monitoring unit and stored in the operation log storage unit, and analyzes the following indicators of data stored in the disk cache; and
An option setting unit that dynamically sets the options necessary for the disk cache to operate according to the analysis results of the operation analysis unit,
The indicators analyzed by the operation analysis unit include read hit rate, write hit rate, time stored in cache, and free space size,
The options set by the option setting unit include data pre-reading and window size. delete The method of claim 1, wherein the operation monitoring unit
Disk cache operation configured to monitor read/write operations performed through the disk cache in order to store information about data read from the disk into the disk cache and operation operation information by which the contents stored in the disk cache are written to the disk. Device. delete The method of claim 1, wherein the option setting unit
A disk cache operating device configured to set a disk cache setting value so that the read hit rate of the disk cache can be improved based on the indicator analyzed by the operation analysis unit and apply the set value to the disk cache. A method of operating the disk cache in a storage system including a disk cache and a disk,
Monitoring the operation log for the disk cache using information transmitted from a metadata storage unit that stores status information and data read error setting information about whether the consistency of data stored in the disk cache is maintained;
storing the monitored operation log;
Analyzing the monitored and stored operation logs to analyze indicators of data stored in the disk cache, where the indicators include read hit rate, write hit rate, time stored in cache, and free space size; and
A disk cache operating method comprising dynamically setting options required for disk cache operation - where the options include data pre-reading and window size - according to the indicator analysis results. delete The method of claim 6, wherein monitoring the operation log
A disk cache that includes monitoring read/write operations performed through the disk cache to store information about data read from the disk into the disk cache and to store operation operation information by which the content stored in the disk cache is written to the disk. How to operate. delete The method of claim 6, wherein dynamically setting the option is
A disk cache operating method comprising setting a disk cache setting value to improve the read hit rate of the disk cache based on the indicator analyzed by analyzing the indicator of the data, and applying the set value to the disk cache. .

Images