KR20140142125A - Computer systerm providing file system - Google Patents

Abstract

Provided is a computer system for providing a file system, which includes a central processing unit (CPU) accessing the file system and a non-volatile memory storing data of the file system and capable of being accessed in byte units. The CPU stores a name space for objects in the file system, into the non-volatile memory. The name space is stored in the form of a data structure employing an in-memory method, in the non-volatile memory. The CPU directly accesses the name space, so the time to search the name space can be reduced.

Description

[0001] COMPUTER SYSTERM PROVIDING FILE SYSTEM [0002]

The following description relates to a computer system that provides a file system, and more particularly to a computer system that provides a file system that includes non-volatile memory.

Data of a file system provided by an existing computer system is stored in a storage device such as a hard disk. Therefore, the name entry included in the namespace of the file system is stored in an object in the file system as an on-disk data structure. If the name entry is stored in the storage device, the data can be preserved even when the computer system is powered off, and the namespace can be managed permanently.

However, when a central processing unit (CPU) processes an operation on an object in the file system, the CPU can not directly access the name entry of the file system stored on the hard disk.

The CPU transfers the stored name entry to the volatile memory as an on-disk data structure, converts the data structure of the name entry into an in-memory data structure, Lt; / RTI >

Also, since the data stored in the hard disk of the computer system can be accessed only in units of blocks, even when a small name entry is moved to the volatile memory, the input / output of the entire block including the name entry occurs in the hard disk. Therefore, the input / output may be overloaded from the hard disk. Conversion between the on-disk type data structure and the in-memory type data structure and the overloading of the block input / output as described above deteriorates the efficiency of the operation processing on the objects in the file system performed by the CPU.

Korean Patent Publication No. 10-2011-0129250 (published on December 1, 2011) discloses a virtual memory system for supporting access of a file system on a main memory including a nonvolatile memory. The disclosed invention discloses a memory system that includes a file system including a main memory and a file access interface including at least one non-volatile memory, and is capable of accessing non-volatile memory through a file access interface in a file system .

The information described above is for illustrative purposes only and may include content that does not form part of the prior art and may not include what the prior art has to offer to the ordinary artisan.

One embodiment may provide a computer system that provides a file system that includes non-volatile memory.

One embodiment may provide a computer system including a non-volatile memory in which a namespace for objects in a file system is stored.

In one aspect, a computer system for providing a file system, the computer system comprising: a central processing unit (CPU) for accessing the file system; and a non-volatile Wherein the CPU stores a namespace for objects in the file system in the non-volatile memory.

The CPU can directly access the namespace stored in the non-volatile memory.

The namespace may include one or more name entries.

The objects may correspond to the one or more name entries, respectively.

The namespace may be a directory in the file system.

The name entry may be an entry in the directory.

The non-volatile memory may be one of STT-MRAM and PCRAM.

The namespace may be stored in the non-volatile memory as an in-memory data structure.

The CPU can directly access the namespace stored in the nonvolatile memory as the in-memory data structure without using the volatile memory of the computer system.

The in-memory data structure may be a chaining hash table.

There is provided a computer system for providing a file system in which a namespace for objects in a file system is stored in a nonvolatile memory so that a namespace can be stored as an in-memory data structure.

There is provided a computer system that provides a file system in which a CPU can directly access a namespace stored in a non-volatile memory, thereby reducing the time to search for a namespace.

1 shows a computer system.
2 illustrates a computer system according to one embodiment.
3 shows a file system and namespace according to an example.

In the following, embodiments will be described in detail with reference to the accompanying drawings. Like reference symbols in the drawings denote like elements.

Various modifications may be made to the embodiments and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the description. It is to be understood, however, that the intention is not to limit the embodiments to the embodiments, but to include all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms used in the examples are used only to illustrate specific embodiments and are not intended to limit the embodiments. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this embodiment belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

In the following description of the present invention with reference to the accompanying drawings, the same components are denoted by the same reference numerals regardless of the reference numerals, and redundant explanations thereof will be omitted. In the following description of the embodiments, a detailed description of related arts will be omitted if it is determined that the gist of the embodiments may be unnecessarily blurred.

1 shows a computer system.

The computer system 100 may include a central processing unit (CPU) 110, a volatile memory 120, and a storage unit 130.

CPU 110 may control the components that computer system 100 includes. In addition, the CPU 110 may process the operations required to control the components that the computer system 100 includes.

For example, the CPU 110 can control the volatile memory 120. [ Also, the CPU 110 may process the operation required to control the volatile memory 120. [

Alternatively, the CPU 110 may access the volatile memory 120. In other words, the CPU 110 can access the data stored in the volatile memory 120.

The volatile memory 120 may be a semiconductor device for storing data. For example, the volatile memory 120 may be a dynamic random access memory (DRAM).

The data may be stored in the volatile memory 120 as an in-memory data structure. The CPU 110 can directly access data stored in an in-memory data structure.

The data stored in the volatile memory 120 may be accessed on a byte-by-byte basis. In other words, the CPU 110 can access the data stored in the volatile memory 120 on a byte-by-byte basis.

The volatile memory 120 may include input / output lines for input / output of data in units of bytes.

When the computer system 100 is powered off, the data stored in the volatile memory 120 may not be preserved. That is to say, the persistence of the data stored in the volatile memory 120 may not be guaranteed.

The storage unit 130 may be a storage device for storing data, or may include a storage device.

The data may be stored in the storage unit 130 as an in-memory data structure. The volatile memory 120 may access the data stored in the storage unit 130. For example, the data stored in the storage unit 130 may be loaded into the volatile memory.

The data stored in the storage unit 130 can be accessed on a block-by-block basis. In other words, the volatile memory 120 can access the data stored in the storage unit 130 on a block basis

The storage unit 130 may include input / output lines for input / output of data on a block-by-block basis.

The storage unit 130 may be a storage device in which data can be accessed, for example, a hard disk.

Alternatively, the storage unit 130 may be a nonvolatile memory, which is a storage device in which data can be accessed, which is accessed on a block-by-block basis. For example, the storage unit 130 may be a flash memory or a solid state drive (SSD) based on a flash memory.

The data stored in the storage unit 130 can be preserved even when the power of the computer system 100 is shut off. That is, the persistence of the data stored in the storage unit 130 can be guaranteed.

The storage unit 130 may include a file system 140.

The file system 140 may be a system for managing objects stored in the storage unit 130 within the operating system of the computer system 100. For example, the file system 140 may be a system for storing and quickly searching for files in a suitable location.

Alternatively, the file system 140 included in the storage unit 130 may be data of a file system provided by the computer system 100. For example, the file system 140 may be data blocks of a file system provided by the computer system 100.

The file system 140 may include the objects stored in the storage unit 130 and the structure of the objects. That is, the file system 140 may include one or more objects.

The file system 140 may include a namespace 150. The namespace 150 may include information about objects in the file system 140.

For example, the namespace 150 may include an inode of the file system 140. The namespace 150 includes information such as name, location, attribute, size, access right, type, owner, and data block pointer of the data block or file included in the objects in the file system 140 .

The namespace 150 may include one or more name entries 160. For example, the name entry 160 may be stored in an object in the file system. In other words, each name entry 160 can be stored in an on-disk data structure within each object in the file system.

The name entry 160 may include some or all of the information of each object included in the namespace 150.

For example, the data structure in which the namespace 150 is stored in the storage unit 130 may be a B-tree data structure. Alternatively, the data structure in which the namespace 150 is stored in the storage unit 130 may be a list data structure.

The CPU 110 may search the namespace 150 of the file system 140 to process operations on the objects in the file system 140.

The CPU 110 may move the name entry 160 to the volatile memory 120 to search for the namespace 150. [

Moving the name entry 160 to the volatile memory 120 may be to load the data into the volatile memory 120. [

The movement of the name entry 160 may be performed through the input / output lines of the storage unit 130. The name entry 160 can be moved through the input / output lines on a block-by-block basis. The name entry 160 is moved to the volatile memory 120 in block units on the input / output lines of the storage unit 130, so that the block input / output may be overloaded.

The name entry 122 moved to the volatile memory 120 may be converted into an in-memory data structure. Namely, the name entry 122 of the on-disk type data structure can be converted into the name entry 124 of the in-memory type data structure. In the volatile memory 120, the name entry 122 of the on-disk type data structure is converted into the name entry 124 of the in-memory type data structure, thereby overloading the data structure type conversion may occur. In addition, the permanence of the name entry 124 may not be guaranteed when the computer system 100 is powered off.

The CPU 110 may obtain information about the object by accessing the name entry 124.

The CPU 110 may perform an operation on the object based on the acquired object information.

The file system, namespace, and name entry are described in further detail in the detailed description which follows.

2 illustrates a computer system according to one embodiment.

The computer system 200 may include a CPU 110, a volatile memory 120, a storage 130, and a non-volatile memory 210, as described above with reference to FIG.

Non-volatile memory 210 may be a semiconductor device for storing data.

The data stored in the nonvolatile memory 210 can be accessed on a byte basis. The nonvolatile memory 210 may include input / output lines for input / output of data in units of bytes.

The data may be stored in the volatile memory 120 as an in-memory data structure.

When the computer system 200 is powered off, the data stored in the nonvolatile memory 210 can be preserved. That is to say, the persistence of the data stored in the nonvolatile memory 210 can be assured.

For example, non-volatile memory 210 may be one of a Spin Transfer Torque Magnetoresistive Random Access Memory (STT-MRAM) and a Phase Change Random Access Memory (PCRAM) .

Non-volatile memory 210 may include a file system 220-1. The file system 220-1 may be data of a file system provided by the computer system 200. [

In other words, non-volatile memory 210 stores data of the file system and can be accessed byte by byte.

The CPU 110 can access the file system 220-1.

The file system 220-1 may correspond to the file system 140 described above with reference to FIG. Or may be part of the file system 140 of FIG.

The file system 220-1 may include a namespace 230. The namespace 230 may correspond to the namespace 150 described above with reference to FIG.

Namespace 230 may be for objects in file system 220-1. In other words, the objects in the file system 220-1 may be sorted or separated according to the namespace 230. [

The storage unit 130 may also include a file system 220-2. The file system 220-2 may correspond to the file system 140 described above with reference to FIG. Alternatively, the file system 220-2 may be part of the file system 140 of FIG.

 The file system 220-1 and the file system 220-2 may be one file system divided. For example, the file system 220-1 and the file system 220-2 may be one in which the file system 140 is stored in the non-volatile memory 210 and the storage unit 130 in a divided manner. That is to say, the CPU 110 can divide the file system 140 into the file system 220-1 and the file system 220-2 and store the file system in the nonvolatile memory 210 and the storage unit 130, respectively.

The CPU 110 determines whether the size of the file system 140 exceeds the size of the storage space of the nonvolatile memory 120 or the size of the file system 140 exceeds the size of the storage space of the non- The file system 140 may be divided and stored in the nonvolatile memory 210 and the storage unit 130. [

Namespace 230 may be for objects in file system 220-2. Namely, the namespace 230 may be a namespace for objects in the file system 220-1 and the file system 220-2.

The file system 220-2 may not include a namespace. That is to say, objects that are classified or separated by the namespace can be stored only in the non-volatile memory 210, not in the storage unit 130.

The CPU 110 may search the namespace 230 to perform operations on objects in the file system 220-1 or 220-2. Unlike the computer system 100 of Figure 1, the computer system 200 may store the namespace 230 for objects in the file system 220-1 or 220-2 in a non-volatile memory 210, .

For example, the CPU 110 may store the namespace 230 for objects in the file system 220-1 or 220-2 in the non-volatile memory 210. [

The namespace 230 may be stored in the non-volatile memory 210 as an in-memory data structure.

As described above with reference to FIG. 1, the CPU 110 can directly access data stored in an in-memory data structure.

That is to say, the CPU 110 can directly access the namespace 230 stored in the non-volatile memory 210. Alternatively, the CPU 110 can directly access the namespace 150 stored in the nonvolatile memory 210 as an in-memory data structure without using the volatile memory 120 of the computer system 200. Here, accessing the namespace 230 may mean accessing an object corresponding to the namespace 230, an object contained within the namespace 230, or an object stored within the namespace 230. The translation of the name entry 160 described above with reference to FIG. 1 and the translation of the data structure into the name entry 124 of the name entry 122 may not be performed.

Since the above movement and the conversion of the data structure are not performed, the overloading in the input / output lines of the storage unit 130 and the overloading due to the conversion of the data structure of the volatile memory 120 can be prevented.

The CPU 110 can search the namespace 230 by directly accessing the namespace 230 stored in the nonvolatile memory 210. [ The CPU 110 directly accesses the namespace 230, thereby improving the processing efficiency of the CPU for the objects.

The namespace 230 may include one or more name entries 232. The name entry 232 may correspond to the name entry 160 described above with reference to FIG.

The name entry 232 may exist in the non-volatile memory 210 as an in-memory data structure. Alternatively, the name entry 232 may correspond to the name entry 124 described above with reference to FIG. Since the name entry 232 exists in the non-volatile memory 210 as an in-memory type data structure, the conversion from the on-disk type data structure to the in-memory type data structure may not be performed .

The in-memory data structure in the non-volatile memory 230 may be a hash table. The address of the data present in the hash table can be determined by the returned key value by applying a predetermined hash function. When a hash function is applied to two different data, if the two returned values are the same, a hash collision may occur. If a hash collision occurs, the data may not be searched correctly. Here, the data may be an object or a name entry 232.

The hash table may be a hashing table based on chaining. By using a chaining based hash table, hash collision can be prevented from occurring. The chaining-based hash table can be a table that links data having the same key value to each other using a chain. Here, the chain may be a unidirectional or bilateral linked-list.

In other words, the in-memory data structure in the non-volatile memory 230 may be a chaining hash table.

The persistence of the namespace 230 and the name entry 232 stored in the nonvolatile memory 210 is also guaranteed when the computer system 200 is powered down as the persistence of the data stored in the nonvolatile memory 210 is ensured .

The technical contents described above with reference to FIG. 1 can be applied as it is, so a more detailed description will be omitted below.

3 shows a file system and namespace according to an example.

The file system 310 may correspond to the file system 140 described above with reference to FIG. Or may be part of the file system 140. For example, file system 310 may be data in a file system.

The file system 310 may include one or more objects 320. 3, the file system 310 may include a namespace 230. [0033]

As described above with reference to FIG. 2, the namespace 230 may include one or more name entries 232. In addition, the namespace 230 may be stored in the non-volatile memory 210 of FIG. 2 as an in-memory data structure.

One or more objects 320 of the file system 310 may correspond to one or more name entries 232. For example, each of the objects 320 may correspond one-to-one to each of the name entries 232.

In other words, one or more objects 320 of the file system 310 may correspond to one or more name entries 232, respectively.

The CPU 110 of FIG. 1 may search the namespace 230 by directly accessing the namespace 230. The CPU 110 may process an operation on the object 320 corresponding to the name entry 232 by searching the namespace 320. [

Each of the objects 320 may include a directory (or folder), a file, a link and a subdirectory of the file system 310, and the like. A directory in the file system may contain one or more entries. Each of the entries may include a file, a soft link and a hard link, a subdirectory.

The namespace 230 may be a directory in the file system 230, and the name entry 232 may be an entry in the directory.

Namely, the namespace 230 may include one or more entries as a directory of the object 320. The name entry 232 may include a file, a soft link, a hard link, and a subdirectory as an entry that the directory of the object 320 contains

The CPU 110 of FIG. 1 may directly access the namespace 230 to handle operations on files, soft links, hard links, and subdirectories contained in the name entry 232 of the namespace 230.

The technical contents described above with reference to FIG. 1 and FIG. 2 can be applied as they are, so that a more detailed description will be omitted below.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

210: Nonvolatile memory
220-1: File system
230: Namespace
232: Name Entry
320: object

Claims (7)

A computer system for providing a file system,
A central processing unit (CPU) for accessing the file system; And
A nonvolatile memory that stores data of the file system and can be accessed in units of bytes;
Lt; / RTI >
Wherein the CPU stores a namespace for objects in the file system in the non-volatile memory. The method according to claim 1,
Wherein the CPU directly accesses the namespace stored in the non-volatile memory. The method according to claim 1,
The namespace comprising one or more name entries,
Wherein the objects correspond to the one or more name entries, respectively. The method according to claim 1,
Wherein the namespace is a directory in the file system and the name entry is an entry in the directory. The method according to claim 1,
Wherein the non-volatile memory is one of STT-MRAM and PCRAM, The method according to claim 1,
The namespace is an in-memory data structure stored in the non-volatile memory,
Wherein the CPU directly accesses the namespace stored in the non-volatile memory as the in-memory data structure without using the volatile memory of the computer system. The method according to claim 6,
Wherein the in-memory data structure is a chaining hash table.

Images