KR20130068590A - Method and metadata server for managing storage device of cloud storage - Google Patents

Abstract

PURPOSE: A method for managing a storage device of a cloud storage and a metadata server thereof are provided to efficiently manage storage devices with different features by managing the storage devices according to the features of the storage devices, allocating the storage devices according to service features, and controlling the data storage of the storage devices. CONSTITUTION: A storage device control unit(101) classifies storage devices included in data servers(200) according to a data processing speed and a capacity and generates storage device groups. A metadata control unit(102) stores metadata including position information of data stored in the storage devices. If a client(300) using the data servers transmits data to be stored in the data servers, the storage device control unit stores the data in a first storage device with the fastest data processing speed and the metadata control unit stores metadata including the position information. [Reference numerals] (100) Metadata server; (101) Storage device control unit; (102) Metadata control unit; (103) Volume control unit; (200) Data server; (300) Client; (400) Network switch

Description

Method and metadata server for managing storage device of cloud storage}

The present invention relates to a method and a metadata server for managing a storage device of cloud storage, wherein the metadata server manages a plurality of storage devices according to characteristics of a plurality of storage devices, and a service characteristic. By separately assigning storage devices and controlling data storage of a plurality of storage devices according to the data to be stored, it is possible to efficiently manage storage devices having different characteristics in cloud storage and to improve service quality. A method and metadata server for managing storage devices.

In the conventional storage environment, data produced by a company has become mainstream, but multimedia data such as blogs, photos, and videos are rapidly increasing due to the rapid development of the Internet technology. In particular, large portal companies that are providing Internet services at home and abroad are newly generating data of several terabytes (TB) to several tens of TB every month. However, the conventional storage structure environment has many problems in terms of storage scalability and ease of management, so it is not easy to apply to such a changing service environment.

Recent technological advances in storage systems or file systems have greatly improved the scalability and performance of storage systems. In terms of the file system structure, some systems are trying to improve the scalability and performance of distributed storage systems by separating the data I / O path of the file and the metadata management path of the file. This structure allows the client system to directly access the storage devices, and can increase storage scalability by avoiding bottlenecks caused by frequent file access. In such a network-based cloud storage environment, client file systems, metadata servers, and data servers communicate over a network to provide input and output of data. To access a specific file, the client first obtains the location information of the block in which the actual data of the file is stored from the metadata server, and then accesses the data server where the block is located and reads the block data and returns it to the client.

However, such a conventional network-based cloud storage structure does not consider the different characteristics of the hundreds or more of the storage devices constituting the storage, there is a problem that the flexible utilization of the storage device to fit the service characteristics is difficult.

An object of the present invention is to manage a plurality of storage devices according to characteristics of a plurality of storage devices in a metadata server, allocate storage devices separately according to service characteristics, and control data storage of the plurality of storage devices according to data to be stored. Accordingly, the present invention provides a method and metadata server for managing storage devices of cloud storage that can efficiently manage storage devices having different characteristics in cloud storage and improve service quality.

According to an aspect of the present invention, there is provided a method for managing storage of a cloud storage, the plurality of data servers including a list of a plurality of storage devices included in the plurality of data servers and the information related to the plurality of storage devices. Transmitting the to the metadata server; Classifying, by the metadata server, the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices; And generating, by the metadata server, a plurality of storage groups for each classification of the plurality of storage devices.

The method for managing storage of the cloud storage comprises: transmitting data to be stored in the plurality of data servers by a client using the plurality of data servers; Controlling the metadata server to store the data in a first storage device having the fastest data processing speed among the plurality of storage devices; And storing, by the metadata server, metadata including location information of the data.

The method for managing storage of the cloud storage comprises: controlling the metadata server to store copy data of the data in a second storage device of the plurality of storage devices; And storing, by the metadata server, metadata including location information of the copy data.

The method for managing the storage device of the cloud storage may include: when the access frequency of data stored in a third storage device among the plurality of storage devices falls below a predetermined setting value, the metadata server causes the third storage device to be used. Controlling the data stored in the storage device to be moved to a fourth storage device having a lower data processing speed and greater capacity than the third storage device; And changing, by the metadata server, location information of metadata of the data moved to the fourth storage device.

The method for managing the storage device of the cloud storage may include: when the access frequency of the data stored in the fifth storage device among the plurality of storage devices rises above a predetermined setting value, the metadata server determines the fifth storage device. Controlling data to be moved to a sixth storage device having a faster data processing speed than the fifth storage device; And changing, by the metadata server, location information of metadata of the data moved to the sixth storage device.

The method for managing a storage device of the cloud storage includes: generating, by the metadata server, a service volume corresponding to a provided service; And allocating one or more of the plurality of storage groups to the service volume according to the characteristics of the provided service.

The metadata server for managing the storage of the cloud storage according to an embodiment of the present invention may classify the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices included in the plurality of data servers. A storage controller configured to generate a plurality of storage groups for each of the plurality of storage devices; And a metadata controller configured to store metadata including location information of data stored in the plurality of storage devices.

When the client using the plurality of data servers transmits data to be stored in the plurality of data servers, the storage controller may transmit the data to a first storage device having the fastest data processing speed among the plurality of storage devices. Control to store, and the metadata control unit may store metadata including location information of the data.

The storage device controller may control to store the copy data of the data in a second storage device of the plurality of storage devices, and the metadata controller may store metadata including location information of the copy data. .

The storage controller may process the data stored in the third storage device than the third storage device when the access frequency of the data stored in the third storage device among the plurality of storage devices falls below a predetermined setting value. The controller may be configured to move to a fourth storage device having a slow speed and a large capacity, and the metadata controller may change location information of metadata of the data moved to the fourth storage device.

The storage device controller may process the data stored in the fifth storage device than the fifth storage device when the access frequency of the data stored in the fifth storage device among the plurality of storage devices increases to a predetermined value or more. The controller may be configured to move to a sixth fast storage device, and the metadata controller may change location information of metadata of the data moved to the sixth storage device.

 The metadata server for managing the storage device of the cloud storage further includes a volume controller for generating a service volume corresponding to a provided service, wherein the storage controller is configured to provide the service volume to the service volume according to the characteristics of the provided service. One or more of the plurality of storage groups may be allocated.

According to an aspect of the present invention, a method and a metadata server for managing a storage device of a cloud storage (cloud storage), the metadata server, a plurality of storage devices according to the characteristics of the plurality of storage devices in the metadata server By managing, allocating storage devices separately according to service characteristics, and controlling data storage of a plurality of storage devices according to data to be stored, it is possible to efficiently manage storage devices having different characteristics in cloud storage and improve service quality. It can provide a method and metadata server for managing the storage of the cloud storage.

According to the method and metadata server for managing the storage of the cloud storage according to an aspect of the present invention, since the storage can be separately managed according to the service characteristics, services that require high-speed processing performance, such as enterprise data analysis service Efficient storage and service management is possible, such as allocating high-speed storage and allocating high-capacity storage for services with large capacity such as data backup.

1 is a diagram schematically illustrating a system for implementing a method for managing storage of cloud storage according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating an operation of a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.
3 is a view for explaining the operation of the volume control unit of the metadata server for managing the storage of the cloud storage according to an embodiment of the present invention.
4 is a diagram illustrating a method of storing data in a service volume generated by a volume control unit of a metadata server for managing a storage device of cloud storage according to an exemplary embodiment of the present invention.
FIG. 5 is a diagram illustrating an operation of storing data by a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.
FIG. 6 is a diagram illustrating an operation of moving data by a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.
7 is a flowchart illustrating a method for managing a storage device of cloud storage according to an embodiment of the present invention.
8 is a diagram illustrating a process of creating and mounting a service volume in a method for managing a storage device of cloud storage according to an embodiment of the present invention.
FIG. 9 is a flowchart illustrating a process of moving old data that is inaccessible in a method for managing a storage device of cloud storage according to an embodiment of the present invention.
10 is a flowchart illustrating a process of moving data in which access increases rapidly in a method for managing a storage device of cloud storage according to an exemplary embodiment of the present invention.

The present invention will now be described in detail with reference to the accompanying drawings. Hereinafter, a repeated description, a known function that may obscure the gist of the present invention, and a detailed description of the configuration will be omitted. Embodiments of the present invention are provided to more fully describe the present invention to those skilled in the art. Accordingly, the shapes and sizes of the elements in the drawings and the like can be exaggerated for clarity.

1 is a diagram schematically illustrating a system for implementing a method for managing storage of cloud storage according to an embodiment of the present invention.

Referring to FIG. 1, a system for implementing a method for managing a storage device of cloud storage according to an embodiment of the present invention includes a metadata server 100, a data server 200, and a client 300 connected to a network switch. The data is exchanged through 400. The metadata server 100 and the data server 200 may be logically divided or separately configured as shown in FIG. 1, but may be included in the same server in other embodiments.

The client 300 mounts a service volume generated by the metadata server 100 so that a user application can access a file system. When a user application accesses a file, first obtain metadata of a file to be accessed from the metadata server 100, and then read data from the data server 200 using location information of data included in the metadata. . Although the client 300 is illustrated as being plural in FIG. 1, this is according to an embodiment, and in another embodiment, the client may be configured as one.

The data server 200 is a server that manages actual data of a file, and may include a data storage manager (not shown) and a data storage device (not shown). In one embodiment, the storage device may include different devices, such as a solid state drive (SSD), a serial attached SCSI (SAS), a serial advanced technology attachment (SATA), and the like, from at least one as much as the server's space permits. It can also be equipped with a storage device.

The metadata server 100 for managing a storage device of cloud storage according to an embodiment of the present invention may be configured to include a storage controller 101, a metadata controller 102, and a volume controller 103. . The configuration of the metadata server 100 shown in FIG. 1 is in accordance with one embodiment, and the blocks shown in FIG. 1 are not all required components, and in other embodiments, some blocks may be added, changed, or deleted. have.

The storage controller 101 is a device for registering and managing all storage devices mounted in the data server 200. The storage controller 101 may classify a plurality of storage groups according to the characteristics of each storage device and manage storage devices corresponding to the storage groups. In an embodiment, the storage controller 101 classifies the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices included in the data server 200, and stores the plurality of storage devices by the classification of the plurality of storage devices. Device groups can be created.

For example, the storage controller 101 may classify a plurality of storage devices into four categories. Table 1 shows a plurality of storage groups classifying the plurality of storage devices.

Storage group Storage type Data processing speed Volume Class A SLC SSD Very fast Very small Class B MLC SSD speed littleness Class C SAS usually usually Class D SATA Slow greatness

Class A is a SLC (Single Level Cell) type SSD device, which has a very small capacity but a very fast data processing speed. Class B is a MLC (Multi Level Cell) type SSD device, which has a small capacity but a high data processing speed. Class C is a SAS device with moderate capacity and storage that provides normal data throughput. Class D is a SATA device, a storage device with large capacity but slow data processing.

By classifying the storage devices as described above, it is possible to allocate and manage devices for managing user data according to each service characteristic, and to mix data of different types of storage devices for more efficient and intelligent data management.

In one embodiment, the real-time data processing service may be assigned a class A as a service that requires a small amount of data but requires high performance input and output of data, such as an e-mail service, an enterprise data processing service, and the like.

In another embodiment, large-scale archiving services may be assigned class D as a service that does not matter speed much like data backup but requires a large amount of storage space.

In another embodiment, a life-cycle service may have recently accessed data such as User Created Contents (UCC) frequently accessed, old data not frequently accessed, but a cumulative amount of service, and recent data may be assigned to class A. Old data can be gradually moved to class D.

Referring back to FIG. 1, the metadata controller 102 manages metadata including file name, size, owner, creation time, and location information of data stored in a plurality of storage devices.

The volume controller 103 is a device that manages a service volume on which the client 300 can mount a file system. In an embodiment, the volume controller 103 may generate a service volume corresponding to the provided service, and allocate one or more of the plurality of storage groups to the service volume according to the characteristics of the provided service.

FIG. 2 is a diagram illustrating an operation of a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.

Although FIG. 2 illustrates a case where two data servers 201 and 202 are provided for clarity, it may be obvious to those skilled in the art that there may be three or more data servers in another embodiment.

The data servers 201 and 202 may be equipped with various types of storage devices such as SSD, SAS, and SATA. In FIG. 2, the data server 201 includes three SLC SSDs 601, 602, and 603. The data server 202 includes two SASs 604 and 605. The number of storage devices mounted on each of the data servers 201 and 202 may vary according to the specifications of the server. In an embodiment of the present invention, the number of storage devices is not limited to the number of storage devices. It may include a storage device.

When the data servers 201 and 202 are initially started, the list of storage devices included in the storage device 201 and 202 and detailed information of each storage device may be registered in the metadata server.

The metadata server collectively manages the storage devices 601 to 605 registered by the data servers 201 and 202 through the storage control unit 101. The storage controller 101 classifies each storage device into four storage groups, such as class A 501, class B 502, class C 503, and class D 504, and classifies individual storage devices into respective storage groups. Manage according to the characteristics of the storage device. In this case, a storage group and a method of managing the storage device may use an algorithm such as a list or an array. In addition to these algorithms, a necessary algorithm may be selectively used as necessary.

3 is a view for explaining the operation of the volume control unit of the metadata server for managing the storage of the cloud storage according to an embodiment of the present invention.

The volume controller 103 is a device for creating and managing a service volume which is a mount unit of a client. When the first service volume is created, the volume controller 103 cannot record data in the form of empty data storage space. Data can be recorded after storage is allocated. That is, the volume controller 103 generates a service volume corresponding to the provided service, and the storage controller 101 allocates one or more of the plurality of storage groups to the service volume according to the characteristics of the provided service. Can be.

The volume control unit 103 may allocate or release the storage device from the storage control unit 101 by four commands 701, 702, 703, and 704 shown in FIG. 3.

ADD DISK 701 is a command for allocating a specific storage device individually to a service volume. Once allocated, data of the service volume can be stored. You can use this command to allocate as many separate storage devices as you need for the service volume.

REMOVE DISK 702 is a command for removing an individual storage device allocated to a service volume. Once removed, the data on this service volume is no longer stored.

ADD CLASS 703 is a command for allocating a specific storage group to a service volume collectively.

REMOVE CLASS 704 is a command for collectively removing a storage group allocated to a service volume.

4 is a diagram illustrating a method of storing data in a service volume generated by a volume control unit of a metadata server for managing a storage device of cloud storage according to an exemplary embodiment of the present invention.

In a real service environment, a plurality of data servers equipped with a plurality of storage devices may be operated. However, in the embodiment illustrated in FIG. 4, four data having different storage devices 606, 607, 608, and 609 are provided. The servers 203, 204, 205, and 206 and three service volumes 801, 802, and 803, each having different characteristics, are assumed to exist and described.

In this case, it is assumed that the service volume 801 is a volume composed of class A and class D, the service volume 802 is a volume composed of class B and class C, and the service volume 803 is a volume composed of class D. The data server 203 including the SLC SSD 606 belongs to class A, and the data server 204 including the MLC SSD 607 belongs to class B, and the data server includes the SAS 608. Assume that 205 belongs to class C, and that data server 206 including SATA 609 belongs to class D.

Since the service volume 801 is a volume composed of class A and class D, data requiring high speed I / O is stored in the SLC SSD 606 of the data server 203 belonging to class A. May be stored in SATA 609 of data server 206 belonging to class D. FIG. That is, data belonging to the service volume 801 is stored only in the data server 203 and the data server 206.

The service volume 802 is a volume consisting of class B and class C. Similar to the service volume 801, the service volume 802 stores data that requires relatively high speed input and output in the MLC SSD 607 of the data server 204 belonging to class B, and has relatively small input / output capacity. Many data can be stored in the SAS 608 of the data server 205 belonging to class C. That is, data belonging to the service volume 802 is stored only in the data server 204 and the data server 205.

The service volume 803 is a volume composed of class D. It can be used when a large amount of data needs to be stored, such as archiving and backup. Data belonging to the service volume 803 is stored only in the data server 206.

FIG. 5 is a diagram illustrating an operation of storing data by a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.

In the embodiment illustrated in FIG. 5, it is assumed that an operation of storing data in a service volume including class A and class D occurs, and the data server 207 is an SLC SSD 610 which is a storage device belonging to class A. It is assumed that the data server 208 includes a SATA 611 that is a storage device belonging to class D.

First, when the client transmits data to be stored in the data server, the storage controller of the metadata server controls to store the data in the first storage device having the fastest data processing speed among the plurality of storage devices. That is, referring to the embodiment of FIG. 5, since the first stored data 1 has a high probability of being accessed again, the storage controller controls the data on the SLC SSD 610 of the data server 207 having a high speed input / output characteristic. (1) can be controlled to store preferentially. The metadata controller of the metadata server may store metadata including location information of data stored in the first storage device.

Metadata server for managing the storage of the cloud storage according to an embodiment of the present invention is not limited to storing the data in one data server to ensure the availability of the file system, another copy to another data server You can copy it. In this case, even if one data server fails, the user application of the client can read the data from the data server to which the data is copied, thereby avoiding the service failure.

To this end, the storage controller may control to store the copy data of the data in a second storage device among a plurality of storage devices. That is, referring to the embodiment of FIG. 5, the storage controller may control to store a copy of the data 1 in the SATA 611 of the data server 208 which is a high capacity storage device. The metadata controller may store metadata including location information of the copy data stored in the second storage device.

According to the storage management policy of a file as an embodiment of the present invention, by storing recently generated files in a high-speed input / output device, not only can it quickly respond to data access requests that occur later, but also to a high-capacity storage device. By maintaining a copy, the availability of the data is also available.

FIG. 6 is a diagram illustrating an operation of moving data by a storage controller of a metadata server for managing storage of cloud storage according to an exemplary embodiment of the present invention.

In the embodiment illustrated in FIG. 6, it is assumed that an operation of moving data in a service volume including class A and class D occurs, and the data server 209 is an SLC SSD 612 which is storage belonging to class A. FIG. It is assumed that the data server 210 includes a SATA 613 that is a storage device belonging to class D.

According to the method for managing the storage of the cloud storage according to an embodiment of the present invention, it is possible to change the storage to be stored according to the access frequency of the data, the data is not accessed for a long time period and the user's access request An example of this rapidly increasing data will be described.

First, when the access frequency of data stored in the third storage device of the plurality of storage devices falls below a predetermined value, the storage controller may slow down data processing speed of the data stored in the third storage device than the third storage device. It can be controlled to move to the fourth storage device having a large capacity. The metadata controller may change location information of metadata of the data moved to the fourth storage device. The predetermined setting value is a setting value that can be determined that the data is inaccessible due to a long period of time when the access frequency falls below the predetermined setting value, and can be set by the user as needed.

Referring to Fig. 6, a description will be given of a method of managing data which is inaccessible due to a long period of time. The metadata in one embodiment is / UCC / 2007 (902), / UCC / 2008 (903), / UCC / 2009 (904), / UCC / according to each year under the / UCC 901 directory. The directory is classified as 2010 (905). According to such a data structure, data existing in the / UCC / 2010 (905) directory has a high probability of being accessed by a user, whereas data existing in the / UCC / 2007 (902) directory has a long period of time and thus has little user access. Have characteristics. When the data is not accessed in this way, when the SLC SSD 612 of the data server 209 belonging to the class A, which is a high-speed low-capacity storage device, there is no space for storing data requiring high-speed input / output. Occurs. Accordingly, the data of the data in the / UCC / 2007 902 directory is moved to and stored in the SATA 613 of the data server 210 belonging to Class D, which is a high-capacity storage device, thereby additionally storing the storage space of the SLC SSD 612. It can be secured.

In addition, when the access frequency of the data stored in the fifth storage device among the plurality of storage devices increases more than a predetermined value, the storage controller may perform faster processing of data stored in the fifth storage device than the fifth storage device. Control to move to the sixth storage device. The metadata controller may change location information of metadata of the data moved to the sixth storage device. The predetermined setting value is a setting value that can be determined that the access request for data is rapidly increased when the access frequency rises above the predetermined setting value and can be set by the user as needed.

Referring to FIG. 6, a method of managing data in which a user's access request increases rapidly will be described. In one embodiment, it is assumed that data is stored in SATA 613 of data server 210. In this case, when the user's access request increases rapidly, the response time for the user's request is increased due to the characteristics of the SATA 613 having low I / O performance. Therefore, the I / O service may be guaranteed by moving to the SLC SSD 612 of the data server 209 belonging to the class A which is a high speed storage device of the files.

On the other hand, as described above, the method of instructing the movement of data according to the access frequency of the data can be made by the internal judgment of the metadata server engine, and it can be performed by the administrator using a separate utility, and additional limitations. Is not.

7 is a flowchart illustrating a method for managing a storage device of cloud storage according to an embodiment of the present invention.

Referring to FIG. 7, if a method for managing storage of cloud storage according to an embodiment of the present invention starts, a list of a plurality of storage devices included in the plurality of data servers and a plurality of data servers and Information related to the plurality of storage devices is transmitted to a metadata server (S11).

The metadata server classifies the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices (S12).

Thereafter, the metadata server generates a plurality of storage groups for each classification of the plurality of storage devices (S13).

According to a method for managing storage of cloud storage according to an embodiment of the present invention, when a client using a plurality of data servers transmits data to be stored in the plurality of data servers, the metadata server may be configured to store the plurality of data servers. The storage device may control to store the data in a first storage device having the fastest data processing speed, and store metadata including position information of the data.

In addition, according to the method for managing the storage of the cloud storage according to an embodiment of the present invention, to ensure the availability of the file system, instead of storing the data in one data server, a separate copy of another data You can copy it to the server. To this end, the metadata server may control to store the copy data of the data in a second storage device among the plurality of storage devices, and store metadata including position information of the copy data.

A method for managing storage of cloud storage according to an embodiment of the present invention is a system for implementing a method for managing storage of cloud storage according to an embodiment of the present invention illustrated in FIGS. 1 to 6. And since the management method is performed similarly to the metadata server, the description in FIGS. 1 to 6 is applied as it is unless otherwise noted, and thus the detailed description will be omitted.

8 is a diagram illustrating a process of creating and mounting a service volume in a method for managing a storage device of cloud storage according to an embodiment of the present invention.

According to the method for managing the storage of the cloud storage according to an embodiment of the present invention, the metadata server generates a service volume corresponding to the provided service, and the plurality of services in the service volume according to the characteristics of the provided service One or more of the storage groups can be allocated. The client may mount a service volume including a plurality of storage groups allocated in this way.

Referring to FIG. 8, according to a method for managing a storage device of cloud storage according to an exemplary embodiment of the present disclosure, the storage device installed in the first data server 211 may be stored in the storage controller 101 of the metadata server. When registering (S21), the device registration is completed according to the classification of the storage device registered by the storage device controller 101 of the metadata server.

The volume control unit 103 generates a service volume for the client 301 to access (S22). At this point, the client 301 can mount the created service volume, but file recording is impossible because there is no space to store data. Therefore, the registered storage device is allocated (S23) so that the client 301 can record the file. In this case, the storage device to be allocated may be selectively performed according to the characteristics of the service volume. If high-speed I / O is required, such as an enterprise data analysis service, class A is allocated. For simple backup service, class D is allocated.

After storage is allocated to the service volume, the client can write the file. The metadata of the file to be recorded is generated (S25) and stored in the volume control unit 103, and the actual data of the file is stored in the storage device of the registered data server 211 (S26).

FIG. 9 is a flowchart illustrating a process of moving old data that is inaccessible in a method for managing a storage device of cloud storage according to an embodiment of the present invention.

According to the method for managing the storage of the cloud storage according to an embodiment of the present invention, since the high-speed storage device has a small capacity characteristics, the input and output by the high-speed storage device as more data is accessed less frequently The quality effect disappears. Therefore, in the case of such data, it moves from the high speed storage device to the low speed high capacity storage device as follows.

To this end, when the access frequency of the data stored in the third storage device among the plurality of storage devices falls below a predetermined value, the metadata server has a higher data processing speed than that of the third storage device. The controller may be controlled to move to the slow and large fourth storage device, and change location information of metadata of the data moved to the fourth storage device. A process of moving old data, which is not frequently accessed, will be described with reference to FIG. 9.

In the embodiment shown in FIG. 9, it is assumed that there is a separate process for monitoring the characteristics of a file in the metadata server. From a user's point of view, it is possible to manage data using a separate utility independent of the metadata server.

Referring to FIG. 9, it is first checked whether the remaining space of the fast storage device is insufficient (S31). If there is a lot of space to store additional data, there is no need to move the data, but if there is not enough space, the old data stored in the high-speed storage device can be moved to the low-speed high-capacity storage device to store the latest files that are frequently accessed on the high-speed storage device. It is necessary to make space.

When the remaining space of the high speed storage device is insufficient, the data of the high speed storage device is analyzed to check whether there is old data that is not accessed, that is, data having an access frequency of less than a predetermined value (S32). If such data exists, it is checked whether a low speed high capacity storage device exists in the volume (S33) and whether there is remaining space in the high capacity storage device (S34). If the space to store the data to be moved exists in the high capacity storage device, the old data is moved from the high speed storage device to the low speed high capacity storage device (S35), and the location information of the metadata of the moved data is changed (S36).

In addition, in the present invention, there is no separate limitation on the data movement between storage devices. You can use Transmission Control Protocol (TCP), User Datagram Protocol (UDP), or a separate method if necessary.

10 is a flowchart illustrating a process of moving data in which access increases rapidly in a method for managing a storage device of cloud storage according to an exemplary embodiment of the present invention.

According to the method for managing a storage device of the cloud storage according to an embodiment of the present invention, when a client access request increases rapidly in the data stored in the low-speed storage device to a high-speed storage device to improve the input and output quality of these files Let's move.

To this end, when the access frequency of the data stored in the fifth storage device among the plurality of storage devices increases more than a predetermined value, the metadata server has a higher data processing speed than the fifth storage device. It is possible to control to move to the sixth fast storage device, and change location information of metadata of the data moved to the sixth storage device. A process of moving data in which such an approach is rapidly increasing will be described with reference to FIG. 10.

In the embodiment shown in FIG. 10, it is assumed that there is a separate process for monitoring the characteristics of a file in the metadata server. From the user's point of view, it is also possible to manage data using a separate utility independent of the metadata server.

Referring to FIG. 10, it is first checked whether data in which access is rapidly increased, that is, data having an access frequency of more than a predetermined set value exists in the low speed storage device (S41). If such data is present, the I / O performance is low due to the characteristics of the low speed storage device, resulting in deterioration of service quality, and it is necessary to move to the high speed storage device to solve this problem. Thereafter, it is checked whether the fast storage device exists among the devices registered in the volume (S42) and whether the remaining space of the fast storage device exists (S43). If the two conditions are met, data that is rapidly approaching is moved from the low speed storage device to the high speed storage device (S44), and the location information of the metadata of the moved data is changed (S45).

In addition, in the present invention, there is no separate limitation on the data movement between storage devices. The TCP method, the UDP method, or a separate method may be used if necessary.

The method for managing the storage device of the above-described cloud storage has been described with reference to the flowchart shown in the drawings. While the above method has been shown and described as a series of blocks for purposes of simplicity, it is to be understood that the invention is not limited to the order of the blocks, and that some blocks may be present in different orders and in different orders from that shown and described herein And various other branches, flow paths, and sequences of blocks that achieve the same or similar results may be implemented. In addition, not all illustrated blocks may be required for the implementation of the methods described herein.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken as limitations. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (12)

A plurality of data servers transmitting a list of a plurality of storage devices included in the plurality of data servers and information related to the plurality of storage devices to a metadata server;
Classifying, by the metadata server, the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices; And
And generating, by the metadata server, a plurality of storage groups according to the classification of the plurality of storage devices. The method according to claim 1,
Transmitting, by a client using the plurality of data servers, data to be stored in the plurality of data servers;
Controlling the metadata server to store the data in a first storage device having the fastest data processing speed among the plurality of storage devices; And
And storing, by the metadata server, metadata including location information of the data. The method according to claim 2,
Controlling the metadata server to store copy data of the data in a second storage device of the plurality of storage devices; And
And storing, by the metadata server, metadata including location information of the copy data. The method according to claim 1,
When the frequency of access of data stored in the third storage device among the plurality of storage devices falls below a predetermined setting value, the metadata server may process data stored in the third storage device more quickly than the third storage device. Control to move to a fourth storage device that is slow and has a large capacity; And
And changing, by the metadata server, location information of metadata of the data moved to the fourth storage device. The method according to claim 1,
When the frequency of access of data stored in the fifth storage device among the plurality of storage devices increases to a predetermined value or more, the metadata server may process the data stored in the fifth storage device more quickly than the fifth storage device. Controlling to move to a sixth fast storage device; And
And changing, by the metadata server, location information of metadata of the data moved to the sixth storage device. The method according to claim 1,
Generating, by the metadata server, a service volume corresponding to a provided service; And
And allocating at least one of the plurality of storage groups to the service volume according to the characteristics of the provided service. A storage controller configured to classify the plurality of storage devices according to data processing speeds and capacities of the plurality of storage devices included in the plurality of data servers, and to generate a plurality of storage groups for each of the plurality of storage devices; And
And a metadata controller configured to store metadata including location information of data stored in the plurality of storage devices. The method of claim 7,
The storage device control unit,
When a client using the plurality of data servers transmits data to be stored in the plurality of data servers, the controller controls to store the data in a first storage device having the fastest data processing speed among the plurality of storage devices.
The metadata control unit,
Metadata server for managing a storage device of the cloud storage, characterized in that for storing the metadata including the location information of the data. The method according to claim 8,
The storage device control unit,
Control to store the copy data of the data in a second storage device of the plurality of storage devices,
The metadata control unit,
Metadata server for managing a storage device of the cloud storage, characterized in that for storing the metadata including the location information of the copy data. The method of claim 7,
The storage device control unit,
When the frequency of access of data stored in the third storage device among the plurality of storage devices falls below a predetermined setting value, the data stored in the third storage device has a lower data processing speed and greater capacity than the third storage device. Control to move to a fourth storage device,
The metadata control unit,
Metadata server for managing the storage device of the cloud storage, characterized in that for changing the location information of the metadata of the data moved to the fourth storage device. The method of claim 7,
The storage device control unit,
When the access frequency of the data stored in the fifth storage device among the plurality of storage devices is higher than a predetermined set value, the sixth storage of the data stored in the fifth storage device faster data processing speed than the fifth storage device To move it to the device,
The metadata control unit,
Meta data server for managing the storage device of the cloud storage, characterized in that for changing the location information of the metadata of the data moved to the sixth storage device. The method of claim 7,
Further comprising a volume control unit for generating a service volume corresponding to the provided service,
The storage device control unit,
And assigning one or more of the plurality of storage groups to the service volume according to the characteristics of the provided service.

Images