WO2016069031A1

WO2016069031A1 - Managing a storage pool

Info

Publication number: WO2016069031A1
Application number: PCT/US2014/070570
Authority: WO
Inventors: Kishore Kumar M; Soundarajan Krishnaswamy; Anand A GANJIHAL
Original assignee: Hewlett Packard Enterprise Development Lp
Priority date: 2014-11-01
Filing date: 2014-12-16
Publication date: 2016-05-06

Abstract

The present subject matter relates to managing a storage pool. In one example, the present subject matter includes identifying, from amongst a plurality of volumes, meta-data service class volumes and data service class volumes based on attributes of volumes. The present subject matter further includes storing a plurality of objects amongst volumes, such that meta-data of objects is stored on meta-data service class volumes and data of objects is stored on data service class volumes, where each object is one of a frequently accessed object and infrequently accessed object. Further, the present subject matter includes determining at least one data storage device, from amongst data storage devices having data service class volumes, such that the data service class volumes store data of infrequently accessed objects, from amongst the objects, and maintaining the at least one data storage device having the data service class volumes in a low power mode.

Description

MANAGING A STORAGE POOL

BACKGROUND

[0001 ] Storage pools are typically used to provide storage space for file systems, databases, applications, and so on. A storage pool may be understood as a coliection of physical data storage devices operating together logically as a unified storage device to store large quantities of data. Examples of data storage devices that may be included in a storage pool are hard disk drives, flash drives, and optical storage media.

[0002] Each of the different types of data storage devices has a capital expenditure, i.e., initial cost of acquiring the device and an operational expenditure, i.e., cost of maintaining the device in an active state, associated with it. The total cost of ownership of a storage pool is a sum of the capital expenditure and operational expenditure associated with each of the data storage devices in the storage pool. BRIEF DESCRIPTION OF DRAWINGS

[0003] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the figures to reference like features and components:

[0004] Figure 1 schematically illustrates a network system implemented for managing a storage pool, according to an example of the present subject matter;

[0005] Figures 2A and 2B illustrate a file system for managing the storage pool, according to an example of the present subject matter;

[0006] Figure 3A illustrates volumes of a second tier of the storage pool, according to an example of the present subject matter;

[0007] Figure 3B illustrates volumes of the second tier of the storage pool, according to another example of the present subject matter; [0008] Figure 4 illustrates a method for managing the storage pool, according to an example of the present subject matter;

[0009] Figure 5 illustrates another method for managing the storage pool, according to an example of the present subject matter; and

[001 0] Figure 6 illustrates a network environment for managing the storage pool, in accordance with an example of the present subject matter.

DETAILED DESCRIPTION

[001 1 ] With significant growth in data volume in recent years, organizations have adopted Information Lifecycle Management (ILM) for managing data right from its creation and initial storage to the time when the data becomes obsolete and is deleted. ILM generally uses Hierarchical Storage Management (HSM) technique that provides for "tiered storage" of the data. Tiered storage involves assignment of different classes of data storage devices to different types of data. These classes can be based on performance requirements, frequency of use, cost, and so on. As an example of tiered storage, hot data or frequently accessed data may be stored on expensive high performance data storage devices while, cold data or infrequently accessed data may be stored on a cheaper low performance data storage device.

[0012] While data tiering reduces capital expenditure associated with the data storage devices, significant amount of power is consumed in maintaining the data storage devices in an active state, resulting in high operational expenditure. In cases where the data storage devices include rotating media devices, such as spinning hard disk drives (HDD), maintaining these devices in the active state results in significantly high power consumption.

[001 3] The present subject matter, describes a system and a method for managing a storage pool. The system and the method provide for putting data storage devices, such as a spin-down capable data storage devices that store infrequently accessed data in a low power mode to reduce power consumed by the data storage devices, thus minimizing the overall expenditure, such as operational expenditure associated with the storage pool. [0014] In accordance with one example implementation, each of a plurality of objects to be stored in a storage pool is classified as one of a required object and a retained object. Required objects comprise hot data that may be frequently accessed while retained objects comprise cold or warm data that may be infrequently accessed and retained for regulatory purposes. Each object includes data and meta-data associated with the data.

[0015] Subsequently, the data storage devices are grouped into at least a first tier and a second tier, based on at least one attribute associated with each of the data storage devices, such that the required objects are stored in the first tier devices and the retained objects are stored in the second tier devices. Examples of the attributes associated with a data storage device include, but are not limited to, a storage capacity, rotation speed, and a spin-down capability. Further, from amongst a plurality of volumes associated with the data storage devices of each of the first and the second tier, meta-data service class volumes and data service class volumes are identified. The meta-data service class volumes store meta-data of the objects and the data service class volumes store data of the objects.

[0016] Accordingly, the required objects are stored in the first tier of the data storage devices, such that meta-data of the required objects is placed in the meta-data service class volumes of the data storage devices in the first tier and data of the required objects is placed in the data service class volumes of the data storage devices in the first tier. Similarly, retained objects are stored in the second tier of data storage devices, such that meta-data of the retained objects is placed in the meta-data service class volumes of the data storage devices in the second tier and data of the retained objects is placed in the data service class volumes of the data storage devices in the second tier.

[0017] Further, from amongst the data storage devices in the second tier, one or more data storage devices associated with the meta-data service class volumes are maintained in an active mode, while one or more data storage devices associated with the data service class volumes may be maintained in a low power mode, in an example implementation, the active mode of a data storage device may be understood as a normal operational mode where data may be read from or written onto the data storage device and wherein the data storage device consumes its rated power. Further, the low power mode may be an idle mode, a standby mode or a power-down mode. As would be understood, a data storage device may be required to be put in an active state from a low power state for a read or write operation to be performed on the data storage device. The data storage devices, when maintained in the active mode, consume more power as compared to the data storage devices maintained in the low power mode. In an example, the low power mode may be based on a spin-down capability of the data storage devices.

[0018] In cases where a request to access a retained object stored in the second tier of data storage devices is received, a data storage device associated with data service class volume that holds the requested retained object, from amongst the second tier of data storage devices, is identified and put into the active mode to move the retained object to a storage device of the first tier.

[0019] Since, a very large percentage of storage space is utilized by data in comparison to meta-data, therefore, by putting the data storage devices, such as spin-down capable data storage devices that stores infrequently accessed data, in the low power mode, power consumption by the data storage devices is reduced. As a result, operational expenditures associated with the storage pool are also minimized.

[0020] The various systems and the methods are further described in conjunction with the following figures. It should be noted that the description and figures merely illustrate the principles of the present subject matter. Further, various arrangements may be devised that, although not explicitly described or shown herein, embody the principles of the present subject matter and are included within its scope.

[0021] The manner in which the systems and the methods for managing a storage poo! are implemented are explained in details with respect to Figure 1 , Figure 2A, Figure 2B, Figure 3A, Figure 3B, Figure 4, Figure 5, and Figure 6. While aspects of described systems and methods for managing the storage pool can be implemented in any number of different computing systems, environments, and/or implementations, the examples and implementations are described in the context of the following system(s).

[0022] Figure 1 schematically illustrates a network system 100 implemented for managing a storage pool, according to an example of the present subject matter.

[0023] As illustrated in the Figure 1 , the network system 100 includes a file system 102. The file system 102 stores and organizes data in a storage pool 104 comprising one or more storage devices, interchangeably referred to as data storage devices.

[0024] in one example, the file system 102 may be implemented as any computing system, such as a desktop, a laptop, a mailing server, and the like, in another example, the file system 102 may be implemented in any network environment comprising a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc. Further, the file system 102 may communicate with a plurality of user devices 106-1 , 106-2, 106-N over a network 108. The user devices 106-1 , 106-2, 106-N, can be referred to as user devices 106, and individually referred to as a user device 106, hereinafter.

[0025] The network 108 may be a wireless network, a wired network, or a combination thereof. The network 108 can also be an individual network or a collection of many such individual networks, interconnected with each other and functioning as a single large network, e.g., the Internet or an intranet. The network 108 can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), and the internet. The network 108 may either be a dedicated network or a shared network, which represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), and Transmission Control Protocol/Internet Protocol (TCP/IP), to communicate with each other. In an example implementation, the network 108 may include a Global System for Mobile Communication (GSM) network, a Universal Mobile Telecommunications System (UMTS) network, or any other communication network that use any of the commonly used protocols, for example, Hypertext Transfer Protocol (HTTP) and Transmission Control Protocol/Internet Protocol (TCP/IP).

[0026] The user devices 106 can include, but are not limited to, desktop computers, laptops, and the like. In an example implementation, one or more users operating the user devices 1 (36 may send a request to the file system 102 for storing one or more objects in the storage devices of the storage poo! 104. In one example, an object may be a file or a block of data. Further, each object includes data and meta-data associated with the data. Once the objects are stored in the storage pool 104, the user devices 106 may also send requests to the file system 102 for accessing the stored objects.

[0027] Although not shown in the Figure 1 , the file system 102 may be coupled to the storage pool 104 over the network 108 or any other network in the network system 100. A storage pool may be understood as a collection of physical storage devices operating together logically as a unified storage device to store large quantities of data. Examples of storage devices that may be included in a storage pool are hard disk drives, flash drives, and optical storage media. As would be understood, the file system 102 perceives each of the storage devices to comprise one or more volumes 1 10 that are logical entities for data storage. Based on the attributes of a storage devices on which a volume 1 1 (3 resides, the volume 1 10 may be a meta-data service class volume or a data service class volume, such that the meta-data service class volume stores meta-data of the objects and the data service class volume stores data of the objects. Examples of attributes include, but are not limited to, a storage capacity, a rotation speed and a spin-down capability. In an example, a storage device associated with meta-data service class volumes may have a higher performance as compared to a storage device associated with data service class volumes. Also, in some example implementations, the storage device associated with meta-data service class volumes may have a smaller storage capacity as compared to the storage device associated with data service class volumes. [0028] In an example implementation of the Figure 1 , the file system 102 includes a data tiering module 1 12. The data tiering module 1 12 identifies at least a first tier 1 14 of storage devices in the storage pool 104 to store required objects and a second tier 1 18 of storage devices in the storage pool 104 to store retained objects. A tier may be understood as a class of storage. For example, a tier may be a collection one or more storage devices of similar performance. Further, the required objects comprise hot data, i.e., data that is frequently accessed and the retained objects comprise cold or warm data, i.e., data that is infrequently accessed and retained based on a retention policy. The retention policy defines one or more terms of retention of an object, such as a retention period for the object.

[0029] For identifying the first tier 1 14 and the second tier 1 16 of the storage devices, the data tiering module 1 12 may determine the attributes of each of the storage devices. As mentioned previously, examples of attributes may include a storage capacity, a rotation speed and a spin-down capability of a storage device, in one example, storage devices that are fast are grouped into the first tier 1 14 and the storage devices that are slow are grouped into the second tier 1 16. In another example, the storage devices that are not spin-down incapable are grouped into the first tier 1 14 and the storage devices that are spin-down capable are grouped info the second tier 1 16. [0030] As described previously, the storage devices include the metadata service class volumes and the data service class volumes, and the storage devices in the first tier 1 14 store required objects and the storage devices in the second tier 1 16 store retained objects. Accordingly, as evident, the first tier 1 14 comprises storage devices associated with the meta-data class volumes holding meta-data of the required objects and the data service class volumes holding data of the required objects. Similarly, the second tier 1 16 comprises storage devices associated with the meta-data class volumes holding meta-data of the retained objects and the data service class volumes holding data of the retained objects. [0031] Further, the file system 102 may also include a service class identification module 1 18 to categorize volumes as meta-data service class volumes and data service class volumes. Accordingly, the storage devices associated with meta-data service class volumes and data service class volumes may be identified by the service class identification module 1 18 amongst the first tier 1 14 as well as the second tier 1 16 of storage devices. Having identified at least one storage device associated with data service class volumes, holding data of the retained objects, from amongst the storage devices in the second tier 1 16; the file system 102 may put the same in the low power mode. Details of functionalities and implementation of the file system 102 have been described in more detail in reference to Figure 2A and 2B.

[0032] Figures 2A and 2B illustrate the file system 102 for managing a storage pool, such as the storage pool 104, according to an example of the present subject matter.

[0033] in accordance with an example implementation illustrated in Figures 2A, the file system 102 includes a processor 202 and a data classification module 204, the data tiering module 1 12, the service class identification module 1 18, and a power management module 206, coupled to the processor 202.

[0034] The processor 202 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Further, functions of the various elements shown in the figures, including any functional blocks labeled as "processors)", may be provided through the use of dedicated hardware as well as hardware capable of executing computer-readable instructions. Among other capabilities, the processor(s) 202 fetches and executes computer-readable instructions stored in memory.

[0035] in operation, the data classification module 204 may identify required objects and retained objects, from amongst the plurality of objects stored in storage devices in the storage pool 104. In an example, the data classification module 204 may receive the plurality of objects from one or more user devices, such as the user devices 106 for storing in the storage pool 104. In one example, the user devices 106 may include applications and databases, and the term "user device" is only for the ease of description. Currently received objects may be considered as required objects or hot data that may be currently in use by one or more of the user devices 106. On the other hand, objects that may have been created by any of the user devices 106 in the past and have remained unaccessed for a predetermined period of time may be identified as retained objects. In one example, the required objects comprise objects that may be frequently accessed while retained objects comprise objects that may be infrequently accessed and retained for regulatory purposes.

[0036] The data classification module 204 may identify the required objects and the retained objects based on one or more predetermined rules, in one example, a predetermined rule for an object may be based on one of time lapse since the object was last accessed and a retention policy associated with the object. The retention policy defines terms of retention of the object, such as a retention period for the object. The retention policy may be defined to meet statutory norms and regulations. For example, for a medical record, a retention period of 10 years may be defined in the retention policy. Accordingly, while the medical record may not be removed from a storage pool until expiration of the retention period, the same may also not be accessed or required by any of the user devices 106 and may be identified as retained data.

[0037] Subsequently, the data tiering module 1 12 may identify, from amongst the storage devices in the storage pool 104, at least a first tier 1 14 of storage devices and a second tier 1 16 of storage devices, based on at least one attribute associated with each of the data storages devices, such that the required objects are stored in the first tier 1 14 and the retained objects are stored are stored in the second tier 1 16. As previously described, the at least one attribute may include a storage capacity, a rotation speed and a spin-down capability, in one example, the storage devices of the first tier 1 14 have a higher speed than the storage devices of the second tier 1 16. [0038] Further, the storage devices in the first tier 1 14 and the second tier 1 16 may he identified to be associated with meta-data service class volumes and data service class volumes by the service class identification module 1 18. The meta-data service class volumes store meta-data of the objects and the data service class volumes store data of the objects. Examples of storage devices of the first tier 1 14 associated with the meta-data service class volumes and data service class volumes, and storage devices of the second tier 1 16 associated with meta-data service class volumes may include flash drives and optical storage media. Further, examples of storage devices of the second tier 1 16 associated with data service class volumes may include rotating media devices that are spin-down capable, such as hard disk drives.

[0039] Further, the service class identification module 1 18 may identify, from amongst the storage devices in the second tier 1 16, storage devices associated with meta-data service class volumes and storage devices associated with data service class volumes. Thereafter, the power management module 206 may maintain the storage devices associated with meta-data service class volumes in an active mode and the storage devices associated with data class volumes in a low power mode. The data storage devices, when maintained in the low power mode, consume less power as compared to the data storage devices maintained in the active mode.

[0040] in an example, the active mode may be understood as a working state in which the storage devices are fully operational, and consumes substantial amount of power. Accordingly, the storage devices in the active mode consume more power, for example, in comparison to the storage devices in the low power mode. The low power mode may include an idle mode, standby mode, or a power-down mode. In the idle mode, the storage devices primarily operate in a lower power and lower power consumption mode. The idle mode may be understood as a power saving mode which may be based on a spin- down capability of the storage devices. During the standby mode, the storage devices are spun down, i.e., rotation speed of the storage devices may be lowered. Further, during the power-down mode or sleep mode, the storage devices are powered off. For example, the sleep mode may use substantially zero power. Although, it has been described that the low power mode may be one of an idle, standby, and power-down mode, the low power mode may include other power saving modes as would be evident to a person skilled in the art from the description provided herein.

[0041] Generally, although retained objects may be stored on cheaper low performance storage devices, such as rotating media devices, significant amount of power is consumed in maintaining the storage devices in the active mode as the storage devices may continuously spin, resulting in high operational expenditure. According to the present subject matter, since data occupies substantial disk space in comparison to meta-data, data of retained objects and meta-data of retained objects are stored separately, such that spin- down capable storage device, such as hard disk drive that hold data of retained objects or infrequently accessed objects may be kept in the low power mode.

[0042] The storage devices having meta-data of infrequently accessed objects are maintained in the active mode to respond to FIND queries for locating or retrieving the objects that may be received from the user devices 108. The storage devices having meta-data of infrequently accessed objects are made active or, in other words, put into the active mode only when a retained object is required by a user device 106 and thus become a required object or when more retained objects are to be stored in the storage device. Accordingly, by maintaining the spin-down capable storage device that stores infrequently accessed data in a low power mode, power consumption by the data storage devices is reduced, thus operational expenditures associated with the storage pool are also minimized.

[0043] Further, accessing meta-data of retained objects for operations like read and write, does not require to switch the second tier storage devices associated with the data service class volumes from the low power mode to the active mode; therefore, the storage pool remains power efficient during ail operations that access meta-data of the retained objects. [0044] In accordance with an example implementation illustrated in Figures 2B, the file system 102, among other things, may include the processor 202, interface(s) 208, a memory 210, modules 212, and data 214, The processor 202, among other capabilities, may fetch and execute computer- readable instructions stored in the memory 210. The memory 210, communicatively coupled to the processor 202, can include a non-transitory computer-readable medium including, for example, volatile memory, such as Static Random Access Memory (SRAM) and Dynamic Random Access Memory (DRAM), and/or non-volatile memory, such as Read Only Memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.

[0045] The interface(s) 208 may include a variety of commercially available interfaces, for example, interfaces for peripheral device(s), such as data input output devices, referred to as I/O devices, storage devices, network devices, and intermediate power devices. The interface(s) 208 may facilitate multiple communications within a wide variety of networks and protocol types, including wired networks and wireless networks.

[0046] Further, the modules 212, amongst other things, include routines, programs, objects, components, and data structures, which perform particular tasks or implement particular abstract data types. The modules 212 may also be implemented as, signal processor(s), state machine(s), logic circuitries, and/or any other device or component that manipulates signals based on operational instructions. Further, the modules 212 can be implemented by hardware, by computer-readable instructions executed by a processing unit, or by a combination thereof. The modules 212 can include the data classification module 204, the data tiering module 1 12, an object storage module 216, the service class identification module 1 18, the power management module 208, and other module(s) 218. The other module(s) 218 may include computer- readable instructions that supplement applications or functions performed by the file system 102. [0047] Further, the data 214 can include rules 220, attributes data 222, and, other data 224. The other data 224 may include data generated and saved by the modules 212 for providing various functionalities to the file system 102.

[0048] As explained above, the data classification module 204 may classify each of a plurality of objects to be stored in storage devices of the storage pool 104 as one of a required object and a retained object. For classifying the objects, the data classification module 204 may retrieve the rules 220 from the data 214. In one example, the rules 220 may be predetermined rules, such as rules based on a retention policy as explained previously.

[0049] Also as explained above, the data tiering module 1 12 may identify different tier of storage devices comprising at least the first tier 1 14 and the second tier 1 16 of storage devices, based on at least one attribute associated with each of the storage devices. In an example implementation, the data tiering module 1 12 may retrieve the attributes data 222 from the data 214. The attribute data 222 may comprise one or more attributes related to the storage devices. The attributes comprise a storage capacity, a rotation speed and a spin-down capability of each of the storage devices. Thereafter, the service class identification module 1 18 may identify the meta-data service class volumes and data service class volumes in the storage devices of both the first tier 1 14 and the second tier 1 16.

[0050] Further, the object storage module 218 may store the required objects in the first tier 1 14 of storage devices and the retained objects in the second tier 1 16 of the storage devices, in an example, the object storage module 216 may stores the meta-data of the required objects in the first tier 1 14 storage devices associated with meta-data service class volumes and data of the required objects in the first tier 1 14 storage devices associated with data service class volumes. Moreover, the object storage module 216 stores the meta-data of the retained objects in the second tier 1 16 storage devices associated with meta-data service class volumes and data of the retained objects in the second tier 1 16 storage devices associated with data service class volumes. In one example, the meta-data may comprise journal and name- space data. In one example, a journal may be a special file thai allows for data recovery of objects, for example in case of power failure. In another example, a journal may be a transaction log comprising records of metadata operation performed by a file system to provide security towards atomicity, consistency, isolation, and durability properties over crashes, hardware failures, or power failure of storage devices in the file system. Further, a name-space data may uniquely identify a set of names so that there is no ambiguity when objects having different origins but same names are mixed together.

[0051] In an example implementation, the service class identification module 1 18 may categorize volumes as meta-data service class volumes and data service class volumes. Accordingly, the storage devices associated with meta-data service class volumes and data service class volumes may be identified by the service class identification module 1 18 amongst the first tier 1 14 as well as the second tier 1 16 of storage devices. Further, the service class identification module 1 18 may identify from amongst the storage devices in the second tier 1 16, at least one storage device associated with meta-data service class volumes and at least one storage device associated with data service class volumes. Based on such identification, the power management module 206 may then maintain the at least one storage device associated with metadata service class volumes in an active mode and the at least one storage device associated with data service class volumes in a low power mode, thereby reducing the power consumed by the at least one storage device associated with data service class volumes.

[0052] Figure 3A and Figure 3B illustrate examples of volumes of a second tier 1 16 of the storage pool 104 of the present subject matter.

[0053] As described previously, the object storage module 216 of the file system 102 may store meta-data of retained objects in storage devices of the second tier 1 16 associated with meta-data service class volumes, and data of retained objects in storage devices of the second tier 1 16 associated with data service class volumes. In one example, a retained object may be a file. Taking the example of a file for the purpose of illustration, for storing the file in the meia-data service class volumes and the data service class volumes, the object storage module 216 may separate meta-data and data of the file. To separate the meta-data and the data of the file, the object storage module 216 may exclude data extents of the file from an index of the file, such that only file attributes are included in the index of the file and data extents are excluded. Subsequently, the object storage module 216 may place the data extents of the file in a storage device associated with the data service class volumes, from amongst the storage devices of the second tier 1 16 and place the file attributes in a storage device associated with meta-data service class volumes.

[0054] in an example implementation illustrated in the Figure 3A, the second tier 1 16 comprises meta-data class volume 302 and data class volume

304. The meta-data class volume 302 includes a plurality of retained object meta-data-1 , retained object meta-data-2,..., retained object meta-data-N.

Moreover, each retained object meta-data is associated with an extent pointer-

1 , For example, the retained object meta-data-1 is associated with an extent pointer-1 , the retained object meta-data-2 is associated with an extent pointer-2, so on and so forth, in an example, an extent pointer may comprise description of an extent.

[0055] For example, the object storage module 216 of the file system 102 may associate an extent pointer with a meta-data of each of the retained objects placed in a storage device associated with the meta-data service class volume 302 to a location of a corresponding data of each of the retained objects in a storage device associated with the data service class volume 304. As may be understood, data of the retained objects may be stored in the data service class volume 304 at disparate locations that may be identified based on an extent map 306. Accordingly, the location of the corresponding data may be stored in the extent map 306.

[0056] in accordance with an example implementation, as illustrated in Figure 3A, the extent map 306 may reside in the data service class volume 304 of the second tier 1 16, while, according to another example implementation, as illustrated in Figure 3B, the extent map 306 may reside in the meta-data service class volume 302 of the second tier 1 16. Details of the extent map 306 have been elaborated subsequently.

[0057] Consider an example of an object that is 64K long and is to be put into data blocks of size 4K each, Further, if the object is stored as two extents of 16K and 4SK, the extent map of the object may store information of where a first block of the 16K extent map is located and a number of consecutive blocks that store data of the object. According to the above example, if a size of a data block is 4K, then the extent map of the object may indicate that next four consecutive blocks contain the data of the object. Moreover, the extent map may also include information about a first block of the 48K extent followed number of consecutive blocks that is twelve blocks, which contain the data of the object.

[0058] As mentioned previously, according to the example implementation illustrated in Figure 3A, the extent map 306 may reside in the data service class volume 304 of the second tier 1 16. Accordingly, as can be seen in the Figure 3A, arrow 308 depicts a location of data corresponding to the retained object meta-data-1 indicated by the extent pointer-1 and arrow 310 depicts a location of data corresponding to the retained object meta-data-2 indicated by the extent pointer-2.

[0059] According to another implementation illustrated in Figure 3B, the extent map 306 may be a part meta-data of the retained objects. Thus, as shown in the Figure 3B, the extent map 306 resides on the meta-data service class volume 302. An extent pointer associated with the metadata of a retained object may point to the extent map 306 residing in the meta-data service class volume 302 and the extent map may in turn point to the location of the data associated with the metadata of the retained object. For example, the arrow 312 depicts the retained object data block- 1 stored in the data service class volume 304 and similarly, arrow 314 depicts the retained object data block- 2 stored in the data service class volume 304.

[0060] The extent map 306, when residing in the meta-data service class volume 302, may be used to respond to various queries relating to locating retained objects, from the user devices or the file system, in a more time effective and power efficient manner. For instance, in response to a query to determine which data service class volume 304 stores data of particular retained objects, the extent map 306 may be used to respond to the query without switching the storage devices associated with the data service class volume 304 from the low power mode to the active mode.

[0061] Further, as described previously, the power management module 206 of the file system 102 may maintain the storage devices associated with the meta-data service class volume 302 in the active mode and the storage devices associated with the data service class volume 304 in the low power mode. In some cases, a user device 106 may wish to access a stored retained object for read/write purposes, in such cases, the object storage module 216 may determine if the storage device of the second tier 1 16, associated with data service class volumes 304, holding the retained object is in an active mode. In case the storage device is in the active mode, the object storage module 216 may move the retained object from the storage device of the second tier 1 16 to a first tier 1 14 of storage device, in an example, when a retained or a cold object is accessed, it becomes a warm or a hot object, and thus is moved to storage devices in the first tier 1 14. However, in case it is determined that the storage device of the second tier 1 16 associated with data service class volume 304 is in a low power mode, the object storage module 216 may move the retained object from the second tier 1 16 to the first tier 1 14 upon switching the storage device of the second tier 1 16 associated with the data service class volume 304 that stored the objects to be accessed by the user device 106 from the low power mode to an active mode.

[0062] in an example, when a user wishes to access a retained object stored in storage devices in the second tier 1 16, both meta-data and data of the retained object move to the storage devices in the first tier 1 14. Since, the storage devices of the second tier 1 16 associated with meta-data service class volume 302 are maintained in the active mode, they can respond to FIND queries and the storage devices of the second tier 1 16 associated with the data service class volume 304 are switched to the active mode from low power mode only when data is to be read or written onto such devices.

[0063] Figure 4 illustrates a method 400 for managing the storage pool, according to an example of the present subject matter and Figure 5 illustrates another method 500 for managing the storage pool, according to an example of the present subject matter.

[0064] The order in which methods 400 and 500 are described are not intended to be construed as a limitation, and some of the described method blocks can be combined in a different order to implement the methods 400 and 500, or alternative methods. Furthermore, the methods 400 and 500 may be implemented in any suitable hardware, computer-readable instructions, or combination thereof.

[0065] The steps of the methods 400 and 500 may be performed by either a computing device under the instruction of machine executable instructions stored on a non-transitory computer readable medium or by dedicated hardware circuits, microcontrollers, or logic circuits. Herein, some examples are also intended to cover computer readable medium, for example, digital data storage media, which are machine or computer readable and encode machine-executable or computer-executable instructions, where said instructions perform some or all of the steps of the described methods 400 and 500.

[0066] With reference to method 400 as depicted in Figure 4, at block 402, the method 400 includes identifying, from amongst a plurality of volumes, meta-data service class volumes and data service class volumes based on attributes of the plurality of volumes. In one example, the attributes of the plurality of volumes comprises a capability of at least one data storage device holding each of the plurality of volumes to be spin-down. In an example, the data tiering module 1 12 of the file system 102 identifies the meta-data service class volumes and the data service class volumes.

[0067] At block 404, the method 400 includes storing a plurality of objects amongst the plurality of volumes, such that meta-data of the plurality of objects is stored on the meta-data service class volumes and data of the plurality of objects is stored on the data service class volumes, where each of the plurality of objects is one of a frequently accessed object and infrequently accessed object. According to an example, frequently accessed objects and infrequently accessed objects, from amongst the plurality of objects, may be identified based on one of time lapse since an object, from amongst the plurality of objects, was last accessed and a retention policy associated with each of the plurality of objects. In an example, the object storage module 216 stores the objects amongst the plurality of volumes.

[0068] At block 406, the method 400 includes determining at least one data storage device, from amongst data storage devices having the data service class volumes, such that the data service class volumes store data of infrequently accessed objects, from amongst the plurality of objects. According to an example, a data storage device, from amongst the data storage devices holding the meta-data service class volumes and the data service class volumes of the first tier, has a higher speed than a data storage device holding the metadata service class volumes and the data service class volumes of the second tier. In an example, the service class identification module 1 18 determines the at least one data storage device associated with data service class volumes that stores data of infrequently accessed objects.

[0089] At block 408, the method 400 includes maintaining the at least one data storage device having the data service class volumes storing data of infrequently accessed objects in a low power mode. The low power mode of the data storage device may be based on the attributes of the plurality of volumes. The low power mode may be one of an idle mode, a standby mode, and a power-down mode. In an example, the power management module 208 may maintain the at least one data storage device having the data service class volumes in the low power mode.

[0070] With reference to method 500 as depicted in Figure 5, at block 502, the method 500 includes identifying required objects and retained objects, from amongst a plurality of objects in a storage pool. Required objects comprise hot data that may be frequently accessed while retained objects comprise cold or warm data that may be infrequently accessed and retained for regulatory purposes. Each object includes data and meta-data associated with the data, in an example, the data classification module 204 of the file system 102 may include the required objects and the retained objects, from amongst the plurality of objects in the storage pool 104.

[0071] At block 504, the method 500 includes identifying a first tier of storage devices in the storage pool to store the required objects and a second tier of storage devices in the storage pool to store the retained objects. In one example, the first tier and the second tier storage devices may be identified based on at least one attribute. Examples of attributes associated with a storage device include, but are not limited to, a storage capacity, rotation speed, and a spin-down capability. According to an example, the service class identification module 1 18 may identify the first tier 1 14 of storage devices and the second tier 1 16 of storage devices.

[0072] At block 506, the method 500 includes storing the required objects in the first tier of storage devices and the retained objects in the second tier of storage devices, such that meta-data of the retained objects is placed on at least one storage device associated with meta-data service class volumes, and the data of the retained objects is placed in at least one storage device associated with data service class volumes, in one example, the data storage devices having data service class volumes have a capability to be spin-down. Further, for placing the meta-data of the objects in the meta-data service class volumes, a pointer may be associated to a data of each object with the corresponding meta-data. According to an example, the object storage module 216 may store the required objects in the first tier 1 14 of storage devices and the retained objects in the second tier 1 16 of storage devices.

[0073] At block 508, the method 500 includes maintaining the at least one storage device associated with meta-data service class volumes in an active mode and the at least one storage device associated with data service class volumes in a low power mode. In an example, the power management module 206 may maintain the at least one storage device associated with meta-data service class volumes in an active mode and the at least one data storage device storing the data service class volumes in a low power mode.

[0074] At block 510, the method 500 includes switching the at least one storage device associated with data service class volumes from the low power mode to the active mode, to move a retained object, from amongst the retained objects stored in second tier of storage devices, to the first tier of storage devices, in one example, the object storage module 216 may switch the storage device, associated with data service class volumes, from low power mode to the active mode to move the retained object to the first tier of storage devices.

[0075] Figure 6 illustrates a network environment 600 for managing the storage pool 104, in accordance with an example of the present subject matter. The network environment 600 may comprise at least a portion of a public networking environment or a private networking environment, or a combination thereof. In one implementation, the network environment 600 includes a processing resource 602 communicatively coupled to a non-transitory computer readable medium 604, hereinafter referred to as computer readable medium 604, through a communication link 606.

[0076] For example, the processing resource 602 can include one or more processors of a computing device for file system journaling. The computer readable medium 604 can be, for example, an internal memory device of the computing device or an external memory device. In one implementation, the communication link 606 may be a direct communication link, such as any memory read/write interface, in another implementation, the communication link 606 may be an indirect communication link, such as a network interface, in such a case, the processing resource 602 can access the computer readable medium 604 through a network 608. The network 608 may be a single network or a combination of multiple networks and may use a variety of different communication protocols.

[0077] The processing resource 602 and the computer readable medium 604 may also be coupled to requested data sources 610 through the communication link 606, and/or to communication devices 612 over the network 608. The coupling with the requested data sources 610 enables in receiving the requested data in an offline environment, and the coupling with the communication devices 612 enables in receiving the requested data in an online environment.

[0078] in one implementation, the computer readable medium 604 includes a set of computer readable instructions, implementing a data tiering module 1 12, a service class identification module 1 18, and a power management module 206. The set of computer readable instructions, referred to as instructions hereinafter, can be accessed by the processing resource 602 through the communication link 606 and subsequently executed to perform acts for managing the storage pool. For discussion purposes, the execution of the instructions by the processing resource 602 has been described with reference to various components introduced earlier with reference to description of Figures 1 , 2A, and 2B.

[0079] On execution by the processing resource 602, the data tiering module 1 12 may group a plurality of storage devices of the storage pool 104 into a plurality of tiers comprising at least a first tier 1 14 and a second tier 1 16, where the first tier 1 14 is to store frequently accessed objects and the second tier 1 16 is to store infrequently accessed objects. The frequently accessed objects and the infrequently accessed objects may be identified from amongst a plurality of objects to be stored in the storage pool 104, based on one of time lapse since an object, from amongst the plurality of objects, was last accessed and a retention policy associated with each of the plurality of objects.

[0080] in an example implementation, the data tiering module 1 12 may determine attributes of each of the plurality of storage devices to group the plurality of storage devices. In one example, the attributes may comprise a storage capacity, a rotation speed and a spin-down capability of each of the plurality of storage devices. The data tiering module 1 12 may further identify, from amongst a plurality of volumes corresponding to the storage devices of the second tier 1 16, a meta-data service class volume 302 to store rnefa-data of the infrequently accessed objects and a data service class volume 304 to store data of infrequently accessed objects. In an example, at least one storage device, from amongst the storage devices of the second tier 1 16, associated with the data service class volume 304 may be a rotating media device that is spin-down capable.

[0081] Thereafter, the service class identification module 1 18 may determine the at least one data storage device, from amongst the data storage devices storing the data service class volumes, such that the at least one data service class volume stores data of infrequently accessed objects, from amongst the plurality of objects. Subsequently, the power management module 208 may put the at least one storage device associated with the data service class volume in a low power mode.

[0082] Although implementations of managing the storage pool have been described in language specific to structural features and/or methods, it is to be understood that the present subject matter is not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed and explained in the context of a few implementations for managing the storage pool.

Claims

I/We claim:

1 . A method to manage storage pool comprising data storage devices having a plurality of volumes, the method comprising:

identifying, from amongst the plurality of volumes, meta-data service class volumes and data service class volumes based on attributes of the plurality of volumes;

storing a plurality of objects amongst the plurality of volumes, such that meta-data of the plurality of objects is stored on the meta-data service class volumes and data of the plurality of objects is stored on the data service class volumes, wherein each of the plurality of objects is one of a frequently accessed object and infrequently accessed object;

determining at least one data storage device, from amongst the data storage devices having the data service class volumes, such that the data service class volumes store data of infrequently accessed objects, from amongst the plurality of objects; and

maintaining the at least one data storage device having the data service class volumes storing the data of infrequently accessed objects in a low power mode.

2. The method as claimed in claim 1 further comprising grouping the plurality of volumes into at least a first tier and a second tier, wherein the first tier comprises meta-data service class volumes and data service class volumes corresponding to frequently accessed objects, from amongst the plurality of objects, and wherein the second tier comprises meta-data service class volumes and data service class volumes corresponding to infrequently accessed objects, from amongst the plurality of objects, and

wherein a data storage device, from amongst the data storage devices holding the meta-data service class volumes and the data service class volumes of the first tier, has a higher speed than the data storage device holding the meta-data service class volumes and the data service class volumes of the second tier.

3. The method as claimed in claim 1 further comprising identifying frequently accessed objects and infrequently accessed objects, from amongst the plurality of objects, based on one of time lapse since an object, from amongst the plurality of objects, was last accessed and a retention policy associated with each of the plurality of objects.

4. The method as claimed in claim 1 , wherein the attributes of the plurality of volumes comprises a capability of the at least one data storage device holding each of the plurality of volumes to be spin-down, and wherein the low power mode is based on the attributes of the plurality of volumes.

5. The method as claimed in claim 1 , wherein the low power mode is one of an idle mode, a standby mode, and a power-down mode.

6. The method as claimed in claim 1 , wherein the storing the meta-data of the plurality of objects in the meta-data service class volumes further comprises, associating a pointer to an extent map associated with data of each of the plurality of objects with the corresponding meta-data, and wherein the data of each of the plurality of objects is stored in the data service class volumes.

7. A file system to manage a storage pool, the file system comprising:

a processor;

a data classification module, coupled to the processor, to identify required objects and retained objects, from amongst a plurality of objects to be stored in the storage pool, based on predetermined rules, wherein the required objects comprise hot data to be stored in the storage pool and the retained objects comprise data other than hot data to be stored in the storage pool;

a data tiering module, coupled to the processor, to identify a first tier of storage devices in the storage pool to store the required objects and a second tier of storage devices in the storage pool to store the retained objects;

a service class identification module, coupled to the processor, to identify at least one storage device associated with meta-data service class volumes and at least one storage device associated with data service class volumes, from amongst the storage devices in the second tier, wherein the meta-data service class volumes are to hold meta-data of the retained objects and the data service class volumes are to hold data of the retained objects; and

a power management module, coupled to the processor, to maintain the at least one storage device associated with meta-data service class volumes in an active mode and the at least one storage device associated with data service class volumes in a low power mode.

8. The file system as claimed in claim 7 further comprising an object storage module coupled to the processor, wherein the object storage module is to:

store the required objects in the first tier of the storage devices and the retained objects in the second tier of the storage devices, such that the metadata of the retained objects is placed in the at least one storage device associated with the meta-data service class volumes, and the data of the retained objects is placed in the at least one storage device associated with the data service class volumes; and

move a retained object, from amongst the retained objects stored in second tier of the storage devices, to the first tier of the storage devices, upon switching the at least one storage device associated with data service class volumes, from amongst the second tier of storage devices from the low power mode to the active mode, to move the retained object,

9. The file system as claimed in claim 8, wherein the object storage module is to associate an extent pointer with a meta-data of each of the retained objects placed in the at least one storage device associated with meta-data service class volumes to a location of a corresponding data of each of the retained objects in the at least one storage device associated with data service class volumes.

10. The file system as claimed in claim 8, wherein the retained object, from amongst the retained objects, is a file and wherein the object storage module is to store file system attributes of the file in the at least one storage device associated with meta-data service class volumes, such that the file system attributes are included in an index corresponding to the file.

1 1 . The file system as claimed in claim 10, wherein the object storage module is to:

exclude data extents of the file from the index of the file; and

place the data extents of the file in the at least one storage device associated with data service class volumes, from amongst the second tier of storage devices.

12, The file system as claimed in claim 7, wherein the meta-data of the retained objects comprises journal and name-space metadata.

13. A non-transitory computer-readable medium comprising instructions for managing a storage pool, executable by a processing resource to:

group a plurality of storage devices of the storage pool into a plurality of tiers comprising at least a first tier and a second tier, wherein the first tier is to store frequently accessed objects and the second tier is to store infrequently accessed objects;

identify, from amongst a plurality of volumes corresponding to the storage devices of the second tier, a meta-data service class volume to store meta-data of the infrequently accessed objects and a data service class volume to store data of infrequently accessed objects,

wherein at least one storage device, from amongst the storage devices of the second tier, associated with the data service class volume is a rotating media device, and wherein the rotating media device is spin- down capable; and

store data of the infrequently accessed objects in the at least one storage device associated with the data service class device volume; and

maintaining the at least one storage device associated with the data service class volume in a low power mode.

14. The non-transitory computer-readable medium as claimed in claim 13 further comprising instructions executable to:

identify the frequently accessed objects and the infrequently accessed objects from amongst a plurality of objects to be stored in the storage pool, based on one of time lapse since an object, from amongst the plurality of objects, was last accessed and a retention policy associated with each of the plurality of objects.

15. The non-transitory computer-readable medium as claimed in claim 13, further comprising instructions executable to determine attributes of each of the plurality of storage devices to group the plurality of storage devices, the attributes comprising a storage capacity, a rotation speed, and a spin-down capability of each of the plurality of storage devices.