Discardable Files

FIELD OF THE INVENTION

The present invention generally relates to storage devices and more specifically to a method and to a device for managing files in a storage device.

BACKGROUND

Use of non-volatile storage devices has been rapidly increasing over the years because they are portable and they have small physical size and large storage capacity. Storage devices come in a variety of designs. Some storage devices are regarded as “embedded”, meaning that they cannot, and are not intended to be removed by a user from a host device with which they operate. Other storage devices are removable, which means that the user can move them from one host device (e.g., from a digital camera) to another, or replace one storage device with another.

The digital content stored in a storage device can originate from a host of the storage device. For example, a digital camera, an exemplary host, captures images and translates them into corresponding digital data. The digital camera then stores the digital data in a storage device with which it operates. Digital content that is stored in a storage device may also originate from a remote source: it can be sent to a host of the storage device, for example, over a data network (e.g., the Internet) or a communication network (e.g., a cellular phone network), and then be downloaded by the host to the storage device. The remote source may be, for example, a service provider or a content provider. Service providers and content providers are collectively referred to hereinafter as “publishers”.

Users of storage devices can willingly download media content and advertisements by requesting the media content or the advertisements from publishers. However, sometimes, publishers, trying to increase their income, send content to users without asking their permission, and sometimes even without the users being aware that such content was downloaded to their storage devices. Content that a publisher sends to users without getting their consent are referred to herein as “unsolicited content”. Oftentimes, unsolicited content is intended to be consumed by users after paying, or after committing to pay, the publisher a fee.

By downloading unsolicited content to users' storage devices publishers hope that users will eventually consume the unsolicited content for a fee, thus increasing their income. Publishers storing unsolicited contents on storage devices without asking users' consent, hoping that the users will consume these contents for a fee, is a concept known in the media publishing field as “predictive consignment”. However, unsolicited content may remain stored in a storage device without the user of the storage device knowing of its existence or wanting to consume it. Storing unsolicited content in a storage device reduces the available (i.e., free) user storage space on the storage device, which is undesirable from the user's point of view. A user may find that there is less space in the storage device for the user's own content (e.g., a music file) because someone else (i.e., some publisher) has taken over part of the storage space on the storage device, or that the user may have to reclaim the storage space so taken by deleting the unsolicited content.

One partial solution to the problem of taking over parts of the user's storage space involves blocking publishers' access to the storage device, such as by blocking the publisher's website. This solution may be acceptable for the users but it is problematic from the publishers' point of view because publishers will make fewer sales and lose a potential income source. Another partial solution to this problem involves publishing content to hosts (i.e., storing content files in storage devices of these hosts) and removing the content when it becomes irrelevant. In other words, the publisher that originated the content removes the stored unsolicited content from the storage device when the content becomes irrelevant. An unsolicited content is regarded as irrelevant if the time for its consumption has lapsed, or when there are indications that the user is not likely to consume it.

There is therefore a need to address the problem with unsolicited files. Specifically, while publishers should be allowed to pursue downloads to storage devices of unsolicited content in the course of conducting their business, these downloads should not have a materially deterring effect on the user experience.

SUMMARY

It would, therefore, be beneficial to be able to store unsolicited files in a storage device for as long as the storage space required to accommodate them in the storage device is not required for user's files, and to remove unsolicited files from the storage device in order to guarantee a minimum size of free storage space for user files. Various embodiments are designed to implement such files management, examples of which are provided herein.

To address the foregoing, files stored, or files to be stored, in a storage device are marked either as non-discardable or discardable in a structure of a file system associated with the storage device. Each marked file has associated with it a discarding priority level. A new publisher's file (i.e., an unsolicited file) is permitted to be stored in the storage device only if storing it in the storage device does not narrow a storage usage safety margin, which is reserved for user files, beyond a desired margin. User files, on the other hand, are allowed to be stored in the storage device even if their storage narrows the storage usage safety margin beyond the desired width. However, in such cases, the desired width of the storage usage safety margin is restored by removing one or more discardable files from the storage device. A discardable file is removed from the storage device if its discarding priority level equals or is higher (or lower, as explained herein) than a predetermined discarding threshold value.

BRIEF DESCRIPTION OF THE DRAWINGS

Various exemplary embodiments are illustrated in the accompanying figures with the intent that these examples not be restrictive. It will be appreciated that for simplicity and clarity of the illustration, elements shown in the figures referenced below are not necessarily drawn to scale. Also, where considered appropriate, reference numerals may be repeated among the figures to indicate like, corresponding or analogous elements. Of the accompanying figures:

FIG. 1 is a block diagram of a storage system according to an example embodiment;

FIG. 2 is a block diagram of a storage system according to another example embodiment;

FIG. 3 is a block diagram of a storage allocator according to an example embodiment;

FIG. 4 is a method for managing files according to an example embodiment;

FIG. 5 is a method for managing the storage of discardable files in a storage device according to an example embodiment;

FIG. 6 is a method for marking one or more unsolicited files in a FAT32-structured file system according to an example embodiment;

FIG. 7 is an exemplary directory area associated with a FAT32 table;

FIG. 8 is a FAT32 table according to an example embodiment;

FIG. 9 is an NTFS table according to an example embodiment;

FIG. 10 is a logical image of a FAT-based file system according to example embodiments; and

FIG. 11 demonstrates files' storage management method in accordance with the present disclosure.

DETAILED DESCRIPTION

The description that follows provides various details of exemplary embodiments. However, this description is not intended to limit the scope of the claims but instead to explain various principles of the invention and the manner of practicing it.

In order to address unsolicited content and related issues, user files are given storage priority over other files, and a storage usage safety margin is maintained to guarantee that priority. A “user file” is a file that a user of a storage device has willingly stored, or has approved its storage in the storage device. For example, a music file that the user downloads to her/his storage device is regarded as a user file. Being requested or approved for storage by the user, user files are regarded as “solicited” files.

The “other files” are referred to herein as “publisher files” and “unsolicited files”. A “publisher file” is a file stored in a storage device without the user requesting it or being aware of it; at least not for a while. The user may not want to use an unsolicited file. Unused unsolicited files tend to consume expensive storage space on the user's storage device. Therefore, according to the principles disclosed herein such files are permitted to be stored in the storage device only if storing them does not narrow the storage usage safety margin. Storage priority is rendered to user files by maintaining a free storage space (i.e., a storage usage safety margin) that will be reserved for future user's files. The storage usage safety margin has to be maintained in order to ensure that user files can be stored in the storage device whenever required or desired.

If for some reason the storage usage safety margin gets narrower than desired, one or more unsolicited files will be removed (i.e., deleted) from the storage device in order to restore the storage usage safety margin. Maintaining the storage usage safety margin guarantees storage space for additional user files if such files are downloaded to the storage device. To this end, unsolicited files are marked as “discardable” in a structure of the storage file system and, if required, removed later to reclaim at least the free storage space required to maintain the storage usage safety margin.

Because the likelihood of the user using the various discardable files may differ from one discardable file to another, each unsolicited file (i.e., each discardable file) is assigned in advance a discarding priority level according to one or more criteria such as the probability of using the file, the probable revenue associated with using the file, the file's size, the file's type, the file's location, the file's age, etc. For example, the discarding priority level may be determined by the potential for revenue. According to another example movie trailers or advertisements would have a higher discarding priority than the actual movie because users usually don't like seeing trailers and advertisements. According to another example, the one or more discardable files that are most likely to be used by the user will be assigned the lowest discarding priority level, which means that such files will be the last file(s) to be removed from the storage device. In other words, the higher the usage probability is of a discardable file the lower the level is of the discarding priority level assigned to that file. If the desired storage usage safety margin is not fully restored even though one or more discardable files were removed, additional discardable files will be removed from the storage device until the desired storage usage safety margin is restored.

Briefly, a file system implements a methodology for storing and organizing computer files. A file system includes a set of abstract data types and metadata that are implemented for the storage, hierarchical organization, manipulation, navigation, access, and retrieval of data. The abstract data types and metadata form “directory trees” through which the computer files (also referred to herein as “data files”, or “files” for simplicity) can be accessed, manipulated and launched. A “directory tree” typically includes a root directory and optional subdirectories. A directory tree is stored in the file system as one or more “directory files”. The set of metadata and directory files included in a file system is called herein a “file system structure”. A file system, therefore, includes data files and a file system structure that facilitate accessing, manipulating, updating, deleting, and launching the data files.

File Allocation Table (“FAT”) is an exemplary file system architecture. FAT file system is used with various operating systems including DR-DOS, OpenDOS, MS-DOS, Linux, Windows, etc. A FAT-structured file system uses a table that centralizes the information about which storage areas are free or allocated, and where each file is stored on the storage device. To limit the size of the table, storage space is allocated to files in groups of contiguous sectors called “clusters”. As storage devices have evolved, the maximum number of clusters has increased and the number of bits that are used to identify a cluster has grown. The version of the FAT format is derived from the number of the table bits: FAT12 uses 12 bits; FAT 16 uses 16 bits, and FAT32 uses 32 bits.

Another file system architecture is known as New Technology File System (“NTFS”). Currently, NTFS is the standard file system of Windows NT, including its later versions Windows 2000, Windows XP, Windows Server 2003, Windows Server 2008, and Windows Vista. FAT32 and NTFS are exemplary file systems with which storage device 100 can be provided.

FIG. 1 shows a typical storage device 100. Storage device 100 includes a storage area 110 for storing various types of files (e.g., music files, video files, etc.), some of which may be user files and others may be publisher files. Storage device 100 also includes a storage controller 120 that manages storage area 110 via data and control lines 130. Storage controller 120 also communicates with a host device 140 via host interface 150. Host device 140 may be dedicated hardware or general purpose computing platform.

Storage area 110 may be, for example, of a NAND flash variety. Storage controller 120 controls all of the data transfers to/from storage area 110 and data transfers to/from host device 140 by controlling, for example, “read”, “write” and “erase” operations, wear leveling, and so on, and by controlling communication with host 140. Storage area 110 may contain, for example, user files and publisher's files, protected data that is allowed to be used only by authorized host devices, and security data that is used only internally, by storage controller 120. Hosts (e.g., host 140) cannot directly access storage area 110. That is, if, for example, host 140 asks for, or needs, data from storage device 100, host 140 has to request it from storage controller 120. In order to facilitate easy access to data files that are stored in storage device 100, storage device 100 is provided with a file system 160.

Storage area 110 is functionally divided into three parts: user area 170, publisher area 180, and free storage space 190. User area 170 is a storage space within storage area 110 where user files are stored. Publisher area 180 is a storage space within storage area 110 where publisher files are stored. Free storage space 190 is an empty storage space within storage area 110. Free storage space 190 can be used to hold a user file or a publisher file. Upon storing a user file in free storage space 190, the storage space holding the user file is subtracted from free storage space 190 and added to user area 170. Likewise, upon storing a publisher file in free storage space 190, the storage space holding the publisher file is subtracted from free storage space 190 and added to publisher area 180. If a user file or a publisher file is removed (i.e., deleted) from storage area 110, the freed storage space is added (it returns) to free storage space 190.

If the size of free storage space 190 permits it, the user of storage device 100 can download a user file from host 140 to storage area 110. The downloaded user file will be stored in free storage space 190 and, as explained above, the storage space holding that file will be subtracted from free storage space 190 and added to user area 170. As explained above, user files have priority over other (e.g., publisher) files, and in order to guarantee that priority, a desired storage usage safety margin is set, and, if required, restored, in the way described below.

Host 140 includes a storage allocator 144 to facilitate restoration of free storage space 190. Storage allocator 144 may be hardware, firmware, software or any combination thereof. In general, storage allocator 144 determines whether a file (e.g., file 142) that is communicated to host 140 is either a user file or a publisher file, and then marks the communicated file accordingly (i.e., as a non-discardable file or as a discardable file).

If storage allocator 144 determines that a file (e.g., file 142) communicated to host 140 is non-discardable, for example because the file is a user file, storage allocator 144 stores the file in storage area 110 in a regular way. As explained above, the storage space within storage area 110 that holds the non-discardable file will be added to, or be part of, user area 170. If, however, storage allocator 144 determines that the file communicated to host 140 is discardable, for example because it is a publisher file, storage allocator 144 marks the file as discardable. If free storage space 190 is larger than the desired storage usage safety margin storage allocator 144 also stores the marked discardable file in free storage space 190, and, as explained above, the storage space within free storage space 190 that holds the discardable file is subtracted from free storage space 190 (i.e., the free storage space is reduced) and added to publisher area 180 (the addition is logically shown as discardable file(s) 182).

As explained above, the likelihood that publisher files may be used by the user may vary from one publisher file to another, which makes a publisher file with the least usage likelihood the first candidate for removal form storage area 110. Therefore, in addition to marking a file as non-discardable or discardable storage allocator 144 assigns a discarding priority level to each discardable file prior, concurrently, or after the discardable file is stored in storage area 110.

By marking files as non-discardable or as discardable, assigning a discarding priority level by storage allocator 144, and by using the file system 160 (or an image thereof) of storage device 100, storage allocator 144 “knows” the number of user files and publisher files in storage area 110, and also their sizes and logical locations within storage area 110. Knowing this information (i.e., the number, sizes and locations of the files), and particularly based on one or more marked files, storage allocator 144 manages storage area 110 and the storage of solicited and unsolicited files in storage area 110. Managing storage area 110, or managing storage of files in storage area 110, may include, for example, restoring a storage usage safety margin by selectively removing one or more files marked as discardable, freeing a storage area by removing all files marked as discardable, and remapping clusters of a file to a lower-performance storage module. Managing storage area 110 or files stored therein may include managing other, additional, or alternative aspects of storage area 110 or files stored therein.

Storage allocator 144 also knows, by the discarding level assigned to each discardable file, the order at which discardable files can or should be discarded (i.e., deleted or removed from storage area 110) in order to restore the free storage space originally reserved for future user files (i.e., to restore the desired storage usage safety margin). Accordingly, if a user wants to store a new user file in storage area 110 but there is not enough free storage space to accommodate that user file (which means that the storage usage safety margin is narrow than desired), storage allocator 144 uses the discarding priority levels assigned to the discardable files to iteratively delete one discardable file after another to regain more free storage space (i.e., to extend free storage space 190) until the desired storage usage safety margin is fully restored. As explained above, a fully restored storage usage safety margin guarantees with high probability that an adequate free storage space is reserved for future user files. Discardable files are removed or deleted from storage device 100 only responsive to receiving a request to store a new user files because it is taken into account that the user may want to use a stored discardable file sometime and, therefore, the discardable file is removed from the storage device only if the storage space accommodating that file is required for the new user file. Storage allocator 144 may be embedded or incorporated into host 140, or it may reside externally to host 140 (shown as dashed box 144′) and to storage device 100.

Storage allocator 144 has a representative image of the file system of, or associated with, storage device 100. Storage allocator 144 uses the storage device's file system image to mark files as non-discardable or as discardable, and to assign a discarding level to each discardable file. Because different file systems have different structures, marking files (i.e., as non-discardable or as discardable) and assigning discarding levels is adapted to the used file system structure, as elaborated in FIGS. 6 through 10, which are described below.

FIG. 2 is a block diagram of a portable storage device 200 according to another example embodiment. Storage controller 220 functions like storage controller 120 and storage allocator 244 functions like storage allocator 144. Storage allocator 244 may be hardware, firmware, software or any combination thereof. Storage allocator 244 internally cooperates with storage controller 220. Whenever storage controller 220 receives from host 240 a storage request to store a file in storage area 210 storage controller 220 informs storage allocator 244 of the storage request and storage allocator 244 marks the file either as non-discardable or discardable in the structure of the file system associated with storage device 200. If storage allocator 244 determines that the new file is discardable storage allocator 244 assigns to the new file a discarding priority level according to the file's usage probability. Then, storage allocator 244 evaluates the current size of free storage space 290 and decides whether one or more discardable files should be removed (i.e., deleted) from storage area 210 in order to make room for the new file. If discardable file or files should be removed from the storage device storage allocator 244 decides which file(s) are the current candidate files for removal. Then, storage allocator 244 notifies storage controller 220 of the discardable files that should be removed from storage area 210 and, responsive to the notification, storage controller 220 removes the discardable file or files indicated by storage allocator 244. In some configurations of portable storage device 200 storage allocator 244 may be functionally disposed between storage controller 220 and storage area 210. In configurations where storage allocator 244 is functionally disposed between storage controller 220 and storage area 210, storage allocator 244 or storage area 210 have to assume some of the functions of storage controller 220. In such configurations storage area 210 is comprised of memory units that communicate at a higher level than flash NAND protocols.

FIG. 3 is a block diagram of a storage allocator 300 according to an example embodiment. Storage allocator 300 includes a memory unit 310, a processor 320, and an interface 330. Memory unit 310 may hold a file system structure, or an image of the file system structure, that is associated with a storage device (e.g., storage device 200 of FIG. 2). Processor 320 manages the file system associated with the storage device. Interface 330 may be adapted to cooperate with a host and with a storage controller of a storage device, as demonstrated in FIG. 1, or only with a storage controller of a storage device, as demonstrated in FIG. 2.

Processor 320 is configured or adapted to receive a request via interface 330 to store a file in a storage area of the storage device, and to mark the file either as discardable or as non-discardable in a structure of the file system associated with the storage device with which storage allocator 300 operates. If interface 330 is functionally attached to storage controller 220 of FIG. 2 (and thus receives, for example, SCSI or wrapped USB/MSC commands rather than file level commands), the received request is at a level that is much lower than a file level. That is, the received request would be a request to store sectors at logical block addresses that, when properly interpreted by a host, would correspond to a file. If storage controller 220 supports the NVMHCI protocol, or a networking file system protocol such as NFS or a similar protocol, storage controller 220 can get file level requests. Therefore, the communication between a storage controller such as storage controller 220 and an interface such as interface 330 is not limited to NVMHCI or to NVMHCI-like implementations. Communication interface 330 may be an integral of storage allocator 300, as shown in FIG. 3.

Processor 320 is further configured or adapted to send the marked file to the storage device, marking the file as discardable includes assigning to the file a discarding priority level. If the file system used by the storage device is FAT-based, processor 320 assigns the discarding priority level to the marked file by setting a corresponding value to m uppermost (i.e., most significant) bits (e.g., m=4) in a FAT corresponding to the marked file. The corresponding value set to the most significant bits in the FAT entry, or the value set to the NTFS directory entry, may be, or it may be, related to an attribute of the file. By “attribute” is meant a metadata tag or some data structure in the header of the FAT table or NTFS table that contains information that pertains to the type of the content stored within the table. “Advertisement”, “premium content”, and “promotional (free) content” are exemplary types of contents that may be stored in the FAT table or in the NTFS table. Alternative criteria for setting discarding levels are, for example, the last accessed files, file sizes, file types, etc.

The number m of the uppermost bits of FAT32 entries dedicated for marking files may be four or less than four because those bits are not used. In addition, the more bits are used the more discarding priority levels can be used. For example, using three bits (i.e., m=3) provides eight (2³=8) discarding priority levels and using four bits (i.e., m=4) provides sixteen (2⁴=16) discarding priority levels (i.e., including discarding priority level “0”, which is assigned to non-discardable files). In other words, processor 320 sets the value of the m uppermost bits to 0 if the marked file is non-discardable or to a value between 1 and 2′″−1 if the marked file is discardable. The discarding priority level indicates the priority at which the marked file can or should be discarded from the storage device. For example, depending on the implementation, the value “1” may denote a file that is either discardable with the lowest priority or with the highest priority, and the value “2′″−1” may respectively denote a file that is either discardable with the highest priority or with the lowest priority.

Processor 320 may assign the discarding priority levels to marked files according to an anticipated usage of the files, as explained above in connection with the likelihood or probability that an unsolicited file is going to be used by the user of the storage device. Processor 320 may update the discarding priority level of the marked file with, or responsive to receiving, each request to store a new file in the storage device. Processor 320 may update the discarding priority level of a given marked file independently from one or more new requests to store a file in the storage device. For example, a file that was previously of a high priority may have its priority lowered after a certain time interval. Processor 320 deletes a file that is stored in the storage device if the file has associated with it a discarding priority level that equals or is greater than a predetermined discarding threshold value. Processor 320 may (re)set the discarding threshold value based on the number of file writes or additions, or depending on the anticipated use of free storage space on the storage device or availability of new publisher files.

Memory unit 310 may hold an assignment table 340 that contains discarding priority levels that processor 320 assigns to files stored in the storage device. In addition, assignment table 340 may hold files' identifiers and information that associates files with the discarding priority levels assigned to the files. Assignment table 340 may additionally hold a discarding threshold value. The information held in assignment table 340 allows processor 320 to identify which discardable file or files can be removed form the storage device in order to restore the desired storage usage safety margin.

Responsive to receiving a request to store a new file in the storage device processor 320 evaluates the size of a free storage space (f) on the storage device and stores the new file in the storage device if the evaluated size of the free storage space on the storage device is larger than a predetermined size or, if it is not larger than the predetermined size, processor 320 searches for one or more discardable files within the storage device that can be deleted and, upon finding such file or files, processor 320 deletes that file or files to extend the current free storage space (f) such that the total size of the extended free storage space equals or is larger than the predetermined size. The discardable file or files can be deleted from the storage device if the discarding priority level associated with the discardable files equals or is greater than a predetermined discarding threshold value (for example between 1 and 15 inclusive, for example 15).

After the free storage space is extended enough processor 320 permits the new file to be stored in the extended free storage space. By “free storage space is extended enough” is meant expanding the free storage space by freeing one occupied storage space after another until the total free storage space can accommodate the new file without narrowing the desired storage usage safety margin mentioned above or, equivalently, until the total size of the extended free storage space equals or is greater then a predetermined size or until all discardable files are removed.

Processor 320 can be a standard off-the-shelf System-on-Chip (“SoC”) device or a System-in-Package (“SiP”) device or general purpose processing unit with specialized software that, when executed, performs the steps, operations and evaluations described herein. Alternatively, processor 320 can be an Application-Specific Integrated Circuit (“ASIC”) that implements the steps, operations and evaluations described herein by using hardware.

FIG. 4 is a method for storing discardable files according to one example embodiment. FIG. 4 will be described in association with FIG. 1. At step 410 host 140 receives a request to store file 142 in storage device 100. At step 420 storage allocator 144 marks the file as “discardable” or as “non-discardable” and sends, at step 430, the marked file to storage controller 120 of storage device 100 (i.e., for storage in storage area 110) if free storage space 190 is sufficiently large. A file is marked also in the sense that a discarding priority level is assigned to the file. At step 440 storage allocator 144 manages storage area 110 (through communication with storage controller 120) or files that are stored in storage area 110 based on the marked file and, optionally, based on one or more files that have already been marked.

FIG. 5 is a method for managing the storage of discardable files in a storage device according to one example embodiment. FIG. 5 will be described in association with FIG. 1. A new file is candidate for storage in storage device 100. Knowing the current image of file system 160 of storage device 100, storage allocator 144 evaluates, at step 510, the current size “f” of free storage space 190 to see whether free storage space 190, whose current size is f, can accommodate the new file (i.e., the file that is candidate for storage). In general, the way storage allocator 144 handles the new file depends on whether the new file is a user file or a publisher file. Therefore, storage allocator 144 determines first whether the new file is a user file or a publisher file.

The New File is a User File

At step 520 storage allocator 144 checks whether free storage space 190 can accommodate the new user file. If free storage space 190 can accommodate the new user file (shown as “Y” at step 520), storage allocator 144 stores, at step 560, the new user file in free storage space 190 regardless of whether the desired storage usage safety margin is narrowed by storing the new user file or not. If the desired storage usage safety margin gets narrower (i.e., relative to the desired storage usage safety margin) after storage allocator 144 stores the new user file in free storage space 190, storage allocator 144 takes no further actions with respect to the storage of the new user file.

If, however, the desired storage usage safety margin gets narrower after storage allocator 144 stores the new user file in free storage space 190, step 550 includes an additional step where storage allocator 144 determines which stored discardable file should be deleted first, which discardable file should be deleted second, and so on, in order to maintain the desired storage usage safety margin. Storage allocator 144 determines which discardable file should be deleted first, which should be deleted second, etc. based on discarding levels that storage allocator 144 assigned to the stored discardable files.

If storage allocator 144 determines at step 520 that free storage space 190 cannot accommodate the new user file (shown as “N” at step 520), storage allocator 144 determines, at step 530, whether free storage space 190 and the storage space consumed by discardable files, when combined, is sufficient for storing the new user file. If the combined storage space is insufficient (shown as “N” at step 530), this means that no matter how many discardable will be deleted the new user file cannot be stored in the “non-user” storage area due to its larger size. If the combined storage space is sufficient (shown as “Y” at step 530), storage allocator 144 searches, at step 540, among stored discardable files which discardable file can be deleted in order to free sufficient storage space for the new user file. Storage allocator 144 searches for these discardable files by using the file system of storage device 100 because, as explained above, storage allocator 144 marks files as non-discardable or as discardable in the file system of the storage device. In addition, the discarding levels assigned by storage allocator 144 to marked files are also embedded into the storage device's file system such that each discarding level is associated with the corresponding marked file.

Upon finding a discardable file (“DF”) that should be discarded first (that file is called hereinafter “DF1”), storage allocator 144 deletes file DF1 in order to add, or to return, its storage space (that storage space is called hereinafter “SP1”) to storage space 190.

Then, at step 550 storage allocator 144 checks whether the extended free storage space 190 (i.e., free storage space 190 plus the last returned storage space, or f+SP1) can accommodate the new user file. If the extended free storage space 190 (i.e., f+SP1) still cannot accommodate the new user file (shown as “N” at step 550) storage allocator 144 iteratively repeats step 550 (the iterations are shown at 555) in order to return an additional storage space to free storage space 190 (i.e., by finding and deleting the next discardable file that should be deleted).

Upon finding the next discardable file with the second highest discarding priority (the next discardable file is called hereinafter “DF2”), storage allocator 144 deletes file DF2 in order to free and add additional storage space (the additional storage space is called hereinafter “SP2”) to free storage space 190. Then, at step 550 storage allocator 144 checks again whether the extended free storage space 190 (i.e., free storage space 190 plus the two last freed storage spaces, or f+SP1+SP2) can accommodate the new file. If the extended free storage space 190 (i.e., f+SP1+SP2) still cannot accommodate the new file (shown as “N” at step 540), storage allocator 144 repeats step 540 one more time in order to find the next discardable file that should be deleted. Storage allocator 144 iterates steps 540 and 550 until the accumulated free storage space 190 can accommodate the new user file (shown as “Y” at step 550). Then, at step 560 storage allocator 144 stores the new user file in storage area 110.

As said above, if the actual storage usage safety margin gets narrower than the desired storage usage safety margin after storage allocator 144 stores the new user file in free storage space 190, step 560 may include an additional step in which storage allocator 144 determines which stored discardable file should be deleted first, which discardable file should be deleted second, etc., in order to restore the desired storage usage safety margin.

The New File is a Publisher File

If the new file is a publisher file, storage allocator 144 stores (at step 560) the new publisher file in storage area 110 only if free storage space 190 can accommodate the new publisher file without narrowing the desired storage usage safety margin. That is, if storing the new publisher file would result in narrowing the desired storage usage safety margin storage allocator 144 may decide not to store the new publisher file in storage area 110. In such a case, storage allocator 144 may refrain from taking any action with respect to that file, and delete no file from the storage device to free storage space for the new publisher file. Alternatively, storage allocator 144 may delete at step 540 one or more higher priority discardable files in order to free storage space for a discardable file that has a lower discarding priority. As stated above, files are marked in, and discarding levels are embedded into, the file system of storage device 100, and the way the files are marked and the discarding levels embedded into the file system depends on, or can be adapted to, the used file system.

FIG. 6 is a method for marking an unsolicited file in a FAT32-structured file system according to one example embodiment. FAT32-structured file systems use clusters. As described above in connection with FAT32-structured file systems, the number of bits that are used to identify a FAT32 cluster is 32. FIG. 6 will be described in association with FIG. 1.

At step 610 m uppermost bits of the 32 bits (where m≦4) of each cluster of the FAT32 are allocated or dedicated for marking files as non-discardable or as discardable, as the case may be, and also for holding a corresponding discarding level for each discardable file. Assigning the discarding level to a file is done by setting a corresponding value to the allocated m bits corresponding to the marked file.

At step 620 storage allocator 144 evaluates the level of likelihood at which the user of storage device 100 will use the unsolicited file. Evaluation of the likelihood of using the file can be implemented in various ways that are known to those skilled in the art of consignment files. For example, the evaluation of the likelihood of using the file may be based on monitoring the location of the person using the storage device, and/or on monitored user's previous experience and preferences. Evaluation of the likelihood of using the file may also be based, for example, on the type of content stored within the FAT table or NTFS table (e.g., “advertisement content”, “premium content”, “promotional (free) content”, etc.). Storage allocator 144 may use alternative or additional criteria to evaluate the likelihood at which the file will be used. For example it may use attributes or characteristics of file(s), which may be, or be associated with, the last accessed file(s), file sizes, file types, etc.

After storage allocator 144 evaluates the level of likelihood at which the user will use the unsolicited file storage allocator 144 assigns, at step 630, a discarding priority level corresponding to the evaluated likelihood level of usage of the unsolicited file. The more likely the unsolicited file is going to be used by the user of storage device 100 the lower is the discarding level.

If m equals four bits, this means that the discarding scale provides 15 discarding levels from 1 (i.e., 0001) to 15 (i.e., 1111). That is, discarding level 0 will be assigned to every non-discardable file, discarding level 1 will be assigned to a discardable file with the lowest discarding priority, and discarding level 15 will be assigned to a discardable file with the highest discarding priority. After storage allocator 144 assigns a corresponding discarding level to the unsolicited file, storage allocator 144 sets, at step 640, a corresponding value between 1 and 15 to the four uppermost bits of the clusters associated with the unsolicited file. If the unsolicited file has associated it two or more clusters, the four uppermost bits in each cluster is set to the same value.

At step 650 it is checked whether the unsolicited file is the last file that needs to be evaluated. If the unsolicited file is not the last file that needs to be evaluated (shown as “N” at step 650) another file is evaluated in the way described above. If the unsolicited file is the last file that needs to be evaluated (shown as “Y” at step 650) the unsolicited file(s) is(are) sent to storage device with the m bits for each whose value was set at step 640.

FIG. 7 is an exemplary directory table 700 associated with a FAT32 table. Directory table 700 is only a partial table used for illustration and as such, table 700 does not show all the fields of a FAT directory entry. Directory area 700 holds particulars of files that are stored in a related file system, such as the files names, files size, and where in a related storage space each file begins. The particulars of the files are held in the following fields. Field 710 holds the Disk Operating System (“DOS”) filenames of the files stored in the related file system, field 720 holds the extension of the files, field 730 holds various attributed of the files, field 740 holds the high 16-bitword of the First Cluster Number (“FCN”) of the files, field 750 holds the low part of the First Cluster Number (“FCN”) of the files, and field 760 holds the size of the files. Each FCN number indicates the first logical cluster where a file may be found.

The first entry of directory area 700 holds information for an exemplary file called “REALFILE” (shown at 770). REALFILE 770 has a file extension “DAT”, its FCN is “0000 0002” (shown at 755), and its size is “0000 24E4”. Numbers in table 700 are shown in hexadecimal values. As part of the standard, attribute values “00” (shown at 780) and “20” (not shown in FIG. 7) refer to a “regular” file, whereas attribute value “02” refers to a file that is hidden in the file system. Filename “\xE5Consign” indicates a deleted file, where “\xE5” means that the value of the first byte of the filename is E5 in hex. By way of example, FCN number 0000 0002 (shown at 755) designates the first cluster of file REALFILE.

FIG. 8 is an exemplary partial FAT32 table 800 according to an example embodiment. FAT32 table 800 is shown as a double-word (“DWORD”) array, and the values are hexadecimal values. Reference numeral 810 designates the type of device holding FAT32 table 800, where “F8” refers to a hard drive. FAT32 table 800 includes 23 clusters that are designated as cluster #1 (shown at 820), cluster #2 (shown at 825), . . . , and cluster #23 (shown at 830). FIG. 8 will be described in association with FIG. 7. A cluster in FAT32 table 800 may be the first cluster of a file, or it may point to the next linked cluster of the file, or it may be an End-of-File (“EOF”) indication.

Referring again to directory area 700, the first FCN of file REALFILE (shown at 770) is “0000 0002” (shown at 755), which points at cluster #2 in table 800 of FIG. 8. As shown in FIG. 8 the value of cluster #2 (i.e., the value “000 0003”) points (shown at 840) at cluster #3, which is the next file's cluster. Likewise, the value of cluster #3 (i.e., “0000 0004”) points at cluster #4, which is the next file's cluster. Cluster #4 has the value “OFFF FFFF” (“F” is the hexadecimal digit that represents the decimal value “15”), where “FFF FFFF” (shown at 850) denotes the file's EOF indication, and the zero value (shown at 860) denotes discarding level 0. File REALFILE, therefore, has associated with it three clusters (i.e., cluster #2, cluster #3, and cluster #4).

As explained above, a discarding level 0 is assigned to non-discardable files. It is noted that the most significant hexadecimal digit of each cluster of a particular file is set to the same discarding priority level that is assigned to that file. For example, file REALFILE has been assigned a discarding level “0” and, therefore, each of the most significant hexadecimal digits of clusters #2, #3, and #4 has that value (i.e., value “0”, the “0” values are underlined). According to another example, the file “E5 Consign” whose FCN is “0000 0005” (as shown in FIG. 7) has been assigned a discarding priority level “1”. Therefore, the most significant hexadecimal digit of each of clusters #5 through 12, which pertain to that file, has the value “1” (for example as shown at 870). In other words, according to the present disclosure the most significant hexadecimal digit (or, equivalently, the four uppermost bits of the clusters associated with a particular discardable file are set to the same value corresponding to the discarding priority level assigned to the particular file. As explained above the number m of the uppermost bits used for indicating the discarding priority level may differ from four (i.e., m≦4).

FIG. 9 is an exemplary partial NTFS table 900 according to an example embodiment. NTFS table 900 holds particulars of files such as the file names, the file sizes, etc. NTFS table 900 includes a data field 910 to hold “regular” data (e.g., data 920) for files that change according to “normal” data flow. According to the present disclosure, NTFS table 900 also includes a “Discarding Information” field 915 for holding, discarding information (e.g., discarding information 930) for each evaluated file. Discarding information field 915 may also include information other than the discarding priority level. For example, discarding information field 915 may include information pertaining to the server that supplied the file and an expiration time after which the file must be discarded. Unlike FAT-based file systems, in NTFS-based file systems the discarding values assigned to discardable files are not limited to a maximum number that is dictated by a set of bits. This means that the range of discarding values can be chosen liberally. For example, discarding values can range from 1 to 25. NTFS is an exemplary non-FAT file system. In general, corresponding discarding values may be set to a data field in a non-FAT based file system entries corresponding to marked files.

FIG. 10 is a logical arrangement of file system 1000 of a storage device according to an example embodiment. A storage allocator (e.g., storage allocator 144 of FIG. 1) may either hold file system 1000 of the storage device with which it operates or an image of file system 1000, or the storage allocator may have an access to file system 1000.

File system 1000 includes a boot section 1010, a FAT 1020 associated with file system 1000, directory tables 1030, a files area 1040, and a discardable files area 1050. FAT 1020 includes a discardable files allocations area 1025 that contains the discarding priority levels of discardable files. Directory tables 1030 include access information for accessing whatever files (i.e., discardable files and/or non-discardable files) are stored in the storage device. Files area 1040 contains the non-discardable files. Index and database area 1045 holds indexes for the discardable files and also metadata that is related to the discardable files. The indexes and metadata held in Index and database area 1045 are used to calculate the discarding levels but they are not required during the actual discarding process. Discardable files area 1050 holds the discardable files.

FIG. 11 demonstrates the files management method according to the present disclosure. FIG. 11 will be described in association with FIG. 1. It is assumed that at time T0 two user files (i.e., files “F1” and “F2”) are initially stored in storage area 110. Because files “F1” and “F2” are user files they are stored in user area 170 and the discarding level assigned to them by storage allocator 144 is zero. Because the total storage capacity of storage area 110 is T (shown at 1110) and files F1 and F2 are stored in storage device 100, the size of the remaining free storage space 190 (see FIG. 1) is f (shown at 1120). It is assumed that a publisher wants to store three unsolicited files in storage area 110. As described above, storage allocator 144 evaluates the size of free storage space 190 (or f at 1120) in storage device 100 in order to determine whether storing the publisher's three unsolicited files in storage area 110 will not narrow a desired storage usage safety margin (shown at 1130) that is reserved for future user's files. If storing publisher's three unsolicited files would narrow storage usage safety margin 1130 (i.e., the desired storage usage safety margin) storage allocator 144 will refrain from storing these files.

In this example storage allocator 144 determines that the publisher's three unsolicited files can be stored in storage area 110 without reducing storage usage safety margin 1130. Therefore, at time T1 storage allocator 144 permits storage controller 120 to store the publisher's three unsolicited files in storage area 110. The three publisher's unsolicited files are designated as “P1”, “P2”, and “P3”. Storage allocator 144 also determines the probability that files P1, P2, and P3 will be used by the user of storage device 100 and assigns a corresponding discarding level to each of these file. Storage allocator 144 then stores the discarding levels assigned to the files in the FAT table, as demonstrated in FIG. 8, or in the NTFS table, as demonstrated in FIG. 9.

At time T2 the user of storage device 100 wants to store in storage area 110 two more files (i.e., files “F3” and “F4”). Storage allocator 144 reevaluates the size of free storage space 190 (or f at 1120) in storage device 100 in order to determine whether there is sufficient storage space in storage area 110 to store the additional files (i.e., files F3 and F4). In this example storage allocator 144 determines that the currently free storage space can accommodate files F3 and F4. Therefore, at time T2 storage allocator 144 permits storage controller 120 to store files F3 and F4 in storage area 110.

Because files F3 and F4 are user files the probability that files F3 and F4 will be used by the user of storage device 100 is irrelevant because user files have storage priority over publisher files regardless of how many times, if at all, the user is going to use files F3 and F4. Accordingly, storage allocator 144 assigns a discarding level “0” to files F3 and F4 and stores the assigned discarding level in the FAT table, as demonstrated in FIG. 8, or in the NTFS table, as demonstrated in FIG. 9.

At time T3 the user of storage device 100 wants to store in storage area 110 another file (i.e., file “F5”). Storage allocator 144 reevaluates the size of free storage space 190 (or f at 1120) in storage device 100 in order to determine whether there is sufficient storage space in storage area 110 to store the additional file (i.e., file F5).

In this example storage allocator 144 determines that the currently free storage space can accommodate file F5. Therefore, at time T3 storage allocator 144 permits storage controller 120 to store file F5 in storage area 110. As shown in FIG. 11, storing user file F5 narrows the storage usage safety margin. That is, the free storage space f in storage area 110 that remains after files F1 through F5 and P1 through P3 are stored in storage area 110 is smaller than storage usage safety margin 1130. Therefore, storage allocator 144 reinstates or restores the storage usage safety margin by removing one of the publisher's files (i.e., P1, P2, and P3). A storage usage safety margin is reinstated or restored by removing (i.e., deleting) one or more publisher files because, as explained above, user files have ultimate storage priority.

As described above, the decision which publisher file or publisher files should be removed from the storage area 110 is made by storage allocator 144 based on the discarding priority level that storage allocator 144 assigned to each stored discardable file.

Turning back to FIG. 11, it is assumed that among the stored publisher files P1 through P3 publisher file P3 was assigned the highest discarding priority level (e.g., 13). Therefore, at time T4 file P3 is removed from storage area 110, thus enlarging the free storage space 190. Because the size of free storage space 190 (or f at 1120) at time T4 is larger than storage usage safety margin 1130, there is no need to remove any more publisher files.

The user of storage device 100 may want to remove one or more user files. At time T5 the user removed two of his files (i.e., files F4 and F5), thus further enlarging free storage space 190. The removal of files F4 and F5 has nothing to do with the size of free storage space 190 or the storage usage safety margin because, as stated herein, regaining free storage space or restoring the storage usage safety margin is done by removing as many discardable files as necessary. It is assumed that a publisher wants to store another unsolicited file in storage area 110. As described above, storage allocator 144 evaluates the size of free storage space 190 (or f at 1120) in order to determine whether storing the publisher's unsolicited file in storage area 110 will not narrow storage usage safety margin 1130. If storing the publisher's the new unsolicited file will narrow storage usage safety margin 1130 storage allocator 144 will refrain from storing that file.

In this example storage allocator 144 determines that the publisher's new unsolicited file (i.e., file “P4”) can be stored in storage area 110 without reducing storage usage safety margin 1130. Therefore, at time T6 storage allocator 144 permits storage controller 120 to store the publisher's file P4 in storage area 110. Storage allocator 144 also determines the probability that file P4 will be used by the user of storage device 100 and assigns a corresponding discarding level to this file. Storage allocator 144 then stores the discarding level assigned to file P4 in the FAT table, as demonstrated in FIG. 8, or in the NTFS table, as demonstrated in FIG. 9. The process of storing new publisher's files and new user files and removing stored files may continue while each time a new file is to be added to storage area 110 storage allocator 144 evaluates the current size of free storage space 190 and determines which publisher file or files (if at all) has/have to be removed from storage area 110.

Assigning a discarding level to a discardable file may be based on user experience or preferences, on Global Positioning System (“GPS”) location of the user, and/or on other criteria. For example, if the user of the storage device seems (based on previous user experience) to like certain types of music, the storage allocator may assign a relatively low discarding priority level (e.g., 3 in a scale of 1 to 15) to a publisher's file if that file contains music that is one of the user's favorite types of music. However, if the publisher's music is disliked by the user (i.e., based on previous user experience), the storage allocator may assign to the related publisher's file a higher discarding priority level (e.g., 12 in a scale of 1 to 15). The criteria used to assign a discarding level to a discardable file may include anticipated usage of the file, anticipated revenue associated with using the file, the file's type, the file's size, the file's location in the storage device, the file's age, and other criteria or parameter as specified herein. Other criteria, whether alone or in combination with any of the criteria mentioned herein, may likewise be used, and the assignment of discarding levels may be done using one or more criterions. In addition, different criteria may be used to assign a discarding level to different discardable files.

In another example, if a publisher wants to send to a user a location-dependent advertisement (i.e., an advertisement relating to a product or service rendered within a specific locality), the storage allocator may assign a discarding priority level to the publisher's advertisement that changes according to the user's changing location. That is, the farther the user gets from a particular location, the higher the discarding level would be, because by getting away from the specific locality it can be assumed that the user is not interested in consuming the product or service rendered at the specific locality.

It is noted that the methodology disclosed herein, of marking files and assigning to them discarding levels in associated file system, may have many useful applications, one of which is restoring a storage usage safety margin to guarantee sufficient storage space for user files. For example, a discarding level assigned to a file may be used to remap file clusters to a lower-performing flash module, or to clear the clusters upon request.

The articles “a” and “an” are used herein to refer to one or to more than one (i.e., to at least one) of the grammatical object of the article, depending on the context. By way of example, depending on the context, “an element” can mean one element or more than one element. The term “including” is used herein to mean, and is used interchangeably with, the phrase “including but not limited to”. The terms “or” and “and” are used herein to mean, and are used interchangeably with, the term “and/or,” unless context clearly indicates otherwise. The term “such as” is used herein to mean, and is used interchangeably, with the phrase “such as but not limited to”.

Having thus described exemplary embodiments of the invention, it will be apparent to those skilled in the art that modifications of the disclosed embodiments will be within the scope of the invention. Alternative embodiments may, accordingly, include more modules, fewer modules and/or functionally equivalent modules. The present disclosure is relevant to various types of mass storage devices such as SD-driven flash memory cards, flash storage devices, non-flash storage devices, “Disk-on-Key” devices that are provided with a Universal Serial Bus (“USB”) interface, USB Flash Drives (““UFDs”), MultiMedia Card (“MMC”), Secure Digital (“SD”), miniSD, and microSD, and so on. Hence the scope of the claims that follow is not limited by the disclosure herein.

	Number	Date	Country
Parent	12336089	Dec 2008	US
Child	13172589		US

Discardable Files

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

US Classifications

International Classifications

Abstract

Description

Claims

RELATED APPLICATION

Continuations (1)