US RE43437 E1
Disk images, and the like, are used to emulate storage volumes for the distribution of computer software. An image of a data storage volume, such as a hard disk drive, is stored in a file having a format that enables it to exhibit a behavior which is the same as the storage volume itself. The image files are accessed by means of an associated driver which can support a variety of different file system protocols, thereby permitting the files themselves to be independent of the format requirements of particular file systems. The data in the image file can be compressed in a manner such that a storage volume being emulated appears to be of a specified size, while actually requiring less space to store its contents. With these properties, the image files can be mounted in the manner of a hardware storage device, and large files can be readily transmitted and downloaded in an electronic format. Since the files are preserved in their original form, end-to-end verification remains possible, to ensure the integrity of the downloaded files. Multiple images can be combined into a single tome, for installation at a remote site using a one-button approach.
1. A system for accessing computer-readable files stored on a source device, by a plurality of target computers comprising:
meansa controller for creating a disk image of the source device, wherein said source device is a physical storage volume on which said computer-readable files to be accessed by said plurality of target computers are located, and for storing said disk image on a storage device that is accessible to said plurality of target computers, wherein said disk image is a virtual representation of said physical storage volume such that it includes volume format information that reflects the format of said physical storage volume, and which enables said disk image to be mounted at each of said plurality of target computers; and
a disk image driver at each of said plurality of target computers, having access to file format information which enables said target computers to read files, which exhibit different file formats, contained on said disk image, and wherein the image driver includes an index that identifies correspondence between address locations in the storage volume and address locations in the disk image.
2. The system of
3. The system of claim 21 wherein said disk image contains a compressed version of data in said files, and wherein said index further includes information pertaining to the manner in which the data was compressed.
4. The system of
5. The system of
6. The system of
7. The system of
8. The system of
a controller for creating a disk image of the source device, wherein said source device is a physical storage volume on which said computer-readable files to be accessed by said plurality of target computers are located, and for storing said disk image on a storage device that is accessible to said plurality of target computers, wherein said disk image is a virtual representation of said physical storage volume such that it includes volume format information that reflects the format of said physical storage volume, and which enables said disk image to be mounted at each of said plurality of target computers; and
a disk image driver at each of said plurality of target computers having access to file format information which enables said target computers to read files, which exhibit different file formats, contained on said disk image, wherein said disk image is stored on the storage device in a compressed read/only format comprising a file which contains compressed versions of chunks of data stored in said physical storage volume, and an index which provides a mapping between logical address blocks in said physical storage volume and addresses of corresponding compressed data in said file.
9. The system of
10. The system of
11. The system of
12. The system of
13. The system of
14. The system of
15. The system of
16. A method for providing a remote computer access to files stored on a source device, comprising the steps of:
creating a disk image of said source device, wherein said source device is a physical storage volume which contains said files to be accessed by said remote computer, and wherein said disk image is a virtual representation of said physical storage volume in that said disk image includes volume format information that reflects the format of said physical storage volume;
generating a script file which includes an identification of said disk image;
launching said script file at said remote computer; and
mounting, at said remote computer, the disk image identified in said script file using a disk image driver that has access to volume format information which is needed to mount files, exhibiting different file formats, on the disk image, and wherein the image driver includes an index that identifies correspondence between address locations in the storage volume and address locations in the disk image.
17. The method of
18. The method of
19. The method of
20. The method of
selectively storing said disk image in a storage medium device in any one of the following disk image file formats:
a read/write format comprising a file which contains a copy of every logical address block in said physic al storage volume, regardless of whether the blocks contain data;
an uncompressed read/only format comprising a file which contains volume information and a copy of only those logical address blocks of said physical storage volume which contain data; and
a compressed read/only format comprising a file which contains compressed versions of chunks of data stored in said physical storage volume, and an index which provides a mapping between logical address blocks in said physical storage volume and addresses of corresponding compressed data in said file.
The present invention is directed to the replication and distribution of electronic files, such as computer software, and more particularly to a system for the creation and manipulation of images of electronic storage media.
With the expanding use of computers in widely varied environments, and more significantly the increasing size of software programs and data files that are employed on those computers, the dissemination of computer software is becoming significantly more complex. Originally, when software programs and data files were of relatively small size, distribution of the programs and files by means of floppy disks was feasible. However, as the processing power of computers continues to expand, and internal storage media such as hard disks offer greater capacity, there is a tendency for software programs to become larger in size. Concomitantly, data files which are processed by those programs, particularly files relating to images and other graphics, have also increased in size. As a result, floppy disks no longer offer a convenient medium for distributing electronic data, due to their limited storage capacity. While a user may be quite willing to install a new software program that is contained on a few floppy disks, the time and effort required to install a large program that occupies a significant number of disks to store all of its contents, e.g., 20 or more disks, becomes intolerable.
For this reason, other media having larger storage capacity, such as CD ROMs, have been employed more recently for the distribution of larger amounts of software. While these other types of media decrease the time and effort required to install the software, they still have certain limitations associated with them. For example, the need to separately package and ship each CD ROM, or the like, requires an appreciable amount of overhead on the part of the software manufacturer. In addition, the storage of the media, both at retail outlets and at the end user's site, represents a growing burden as the number of programs and data files continues to expand.
To overcome these limitations associated with the limited size of transportable storage media and the need to accommodate or dispose of the media, it is desirable to distribute software, including both programs and data files, in an electronic format. Currently, installer programs are available which provide a user with an opportunity to install a program on a computer from a remote site. In one type of installer, a large number of bundled files are downloaded from the remote site in a compressed format, and then expanded on the user's computer. From this large number of files, the user picks the few that are desired, and they are installed on the user's computer system. From the standpoint of user convenience, this approach is less than ideal, because of the time spent downloading a number of unnecessary files, as well as the effort required by the user to select the needed files. In contrast to this approach, a “one-button” installation system is preferable. In such an approach, the user is only required to perform a single action, e.g., select a single button in a graphical user interface, to have the appropriate files downloaded and installed.
Typically, in a one-button installer, all required files are compressed into a single file known as a “tome.” The tome is provided together with an installer program and a script file that is downloaded to the user's site. The installer program, together with the information contained in the script file, expands the tome back into the individual files, and installs the files onto the user's computer system. To perform the installation, the user is only required to select the installer program to run. All other actions are carried out automatically thereafter.
While the latter approach offers the convenience of a one-button installation system, to date it has been limited in the types of programs with which it can be employed. More specifically, due to restrictions associated with the number of resources that can be handled in a tome, it is not possible to use the one-button installer system for programs having large files, or a large number of files, such as operating system software.
Furthermore, since installer programs operate on files individually, they are not able to ensure the integrity of files installed on the user's system. More particularly, since each individual file is compressed and expanded during the installation process, it is not possible to provide end-to-end verification of each file. For example, the original file may have a checksum value tagged to it, to provide for verification of the integrity of the file when it is copied. However, as part of the process of compressing the file, the checksum value is not always preserved. For example, the file might be renamed. Consequently, when the file is expanded and installed, it is not impossible to employ the checksum value to verify the integrity of the file.
In accordance with the present invention, the foregoing limitations associated with the distribution of computer software are addressed through the use of disk images, and the like, to emulate storage volumes. An image of a data storage volume, such as a hard disk drive, is stored in a file having a format that enables it to exhibit a behavior which is the same as the storage volume itself. The image files are accessed by means of an associated driver which can support a variety of different file system protocols, thereby permitting the files themselves to be independent of the format requirements of particular file systems. The data in the image file can be compressed in a manner such that a storage volume being emulated appears to be of a specified size, while actually requiring less space to store its contents. With these properties, the image files can be mounted in the manner of a hardware storage device, and large files can be readily transmitted and downloaded in an electronic format. Since the files are preserved in their original form, end-to-end verification remains possible, to ensure the integrity of the downloaded files. Multiple images can be combined into a single image file tome, for installation at a remote site using a one-button approach.
The foregoing features of the invention, as well as the advantages provided thereby, are explained in greater detail hereinafter with reference to exemplary embodiments illustrated in the accompanying drawings.
To facilitate an understanding of the present invention, it is described hereinafter with reference to specific examples relating to the downloading and installation of software on computers. For example, the installation may take place at the time of computer manufacture, or occur in a private context via an enterprise network or a world-wide network, such as the internet. It will be appreciated, however, that these embodiments are merely exemplary, and do not constitute the only practical applications of the invention. Rather, the invention can be successfully utilized in a variety of different situations in which it is desirable to replicate and/or disseminate software in an electronic format.
Generally speaking, the present invention is based upon the use of storage media images to facilitate the dissemination and access to large quantities of data. In the context of the present invention, a disk image is a file which is located on any suitable storage medium and which has properties that cause it to appear to computer system resources as if it were a physical storage device itself, e.g. a hard disk drive. In one aspect, the present invention is directed to a multiple-format disk image, in which the contents of a storage medium can be stored in any one of a variety of formats that provide the most efficient use of backing storage commensurate with the needs of the user. In another aspect, the present invention is directed to a disk image device driver which allows disk images of any of a variety of formats to be mounted and used in the same manner as any type of physical storage device.
One environment in which the features of the present invention are particularly desirable is the downloading of software onto computers. For example, most new computers are typically sold with bundled software, in which a number of programs are installed on the storage media, e.g., hard disk, of the computer. To this end, the software must be downloaded and installed at the manufacturing site. In one approach, this can be accomplished through the use of a disk image file. Referring to
The creation of the image is carried out by downloading the software bundle to the image drive and verifying its integrity, for example, by means of a checksum operation. Thereafter, a target computer 10 is connected to the controller and powered up. Preferably, the computer 10 is connected to a boot drive 16, from which the computer begins a startup operation. At a particular point in the boot process, the computer communicates with the controller over a serial cable 18. In response to a signal from the computer 10, the controller 12 downloads the software bundle onto the internal disk drive of the computer 10. This procedure is accomplished by means of a block-to-block transfer from the image drive 14 to the target drive of the computer 10 over a parallel bus, as illustrated by the arrow 20 in
In the arrangement depicted in
The mounting of a storage medium refers to the process by which information is provided to the file management facility of a computer's operating system, so that the computer can access information on the storage medium. In the following discussion, the term “volume” will be used to refer to different types of storage devices. In essence, a volume can be any piece of a storage medium, such as a disk, which is formatted to contain files. A volume can be an entire disk, or only part of a disk. For example, a floppy disk might comprise a single volume. A larger storage device, such as a hard disk or a file server, can be divided into many different volumes, or partitions, each of which can be formatted in a different manner.
Typically, information in a volume is organized in the form of logical blocks, each of which contains a predetermined amount of information, e.g., 512 bytes. One example of the organization of the logical blocks in a storage volume is shown in
The volume information in logical block 2 comprises various fields that are used by the operating system's file management facility, such as the volume name, it size, and the number of files on the volume. This information is initially created when the volume is initialized, or formatted, and modified thereafter whenever the file management system writes information to the volume. Each time a volume is mounted, the file manager reads the volume information from the logical block and stores it in a predetermined area of the computer's working memory, e.g. its RAM. Once the file manager has retrieved and stored the volume information, the volume is considered to be mounted, and the file manager can access information contained in the remaining blocks of the volume. A volume becomes unmounted when the file manager releases the memory that was used to store the volume information.
Typically, each volume that is mounted in a computer system represents a different physical device, e.g. a floppy disk, or a portion of a physical device, such as a partition on a disk drive. In contrast, a disk image is not associated with any particular physical device. Rather, it comprises a file in some storage medium, referred to hereinafter as the backing store, and is capable of behaving in the manner of a physical device. Thus, the file itself comprises logical blocks containing the information that is normally associated with a physical volume, i.e. system startup information, volume information, and a volume bitmap, in addition to the actual data of the file itself.
In operation, application programs running on a computer communicate with the storage volumes by means of associated drivers. For example,
In a similar manner, when a disk image is mounted, a driver is employed to carry out the task of providing data to the image file and retrieving data therefrom. As shown in
In accordance with one aspect of the present invention, a disk image can have any one of a number of different formats. Three such formats, namely uncompressed read/write, uncompressed read/only, and compressed read/only, are illustrated in
If there is no need to write information to the disk image, it is preferable to store it in a read/only format which saves storage space. In this format, the size of the disk file is commensurate with only the used portion of the source volume. Referring to
To obtain the greatest savings in storage space, the compressed read/only format is employed. In the creation of a compressed read/only image file, the original data is divided into “chunks” of any suitable size. In the example of
Associated with the compressed image file is an index resource 60. This index comprises a mapping of the logical blocks of the source volume 52 to byte ranges in the compressed image file 58. The index also contains information which identifies a particular compression algorithm that was used on the associated chunk of data, and perhaps also the sizes of the chunks, to thereby provide information on the manner in which to decompress the data from the file 58.
The index 60 can also be employed in connection with the other disk image file formats. For example, in the case of the uncompressed read/only format of
Another format for a disk image is a compressed read/write format. In this format, the user can specify a container of a particular size, e.g , a 1-Gigabyte volume. A disk image file is then created with the appropriate volume information pertaining to such a volume. When initially created, the disk image could be empty, i.e. contain no file data. When a command is generated to write data to the disk image, the data is segmented into suitably sized chunks, if necessary, and each chunk is compressed as appropriate. Each compressed chunk of data is then added to the disk image file, and the appropriate mapping and compression data is entered into the index 60. Thus, the disk image file grows as data is written to it, up to the limit of its specified capacity.
Referring again to
Disk image files of the type described above, which emulate the entire contents of the storage volume, can be employed in the electronic dissemination and replication of software.
Several advantages are associated with the creation of disk images to emulate mountable storage volumes and disseminate software. For example, by creating an image of an entire volume, the individual files themselves remain intact. As such, it is possible to employ a checksum technique, or similar such type of data integrity check, to provide end-to-end verification that files have been successfully transferred. More particularly, the disk image 66 contains all of the contents of a source volume (not shown). When the disk image 66 is to be created, the network server 62 can calculate a checksum for the source volume, and store this value in the volume information, which is then included in the disk image file 66. Subsequently, when a remote client computer copies the contents of the disk image file to its own disk drive 74, it can run a checksum on the copied data to determine whether the same value is obtained.
Another advantage stemmning from the use of images of entire storage volumes is the fact that the dissemination of files can be independent of the file systems themselves. For example, with reference to
Another advantage associated with the distribution of software via disk images lies in the reduced bandwidth requirements for disseminating the images over networks. Typically, most network communications are carried out through various layers that comprise distinct levels of capabilities, or services, that build upon one another. One well-known layer architecture is the OSI reference model, which employs seven layers to describe network activities. These seven layers are depicted in the block diagram of
Typically, when a file is transferred from one computer to another via their respective file managers, such communications take place through all seven layers of the network architecture. However, since a disk image functions as a physical device on the computer system, only the lowest level protocols need to be employed to communicate over the network. The services of the higher level layers are not needed. For example, transfer of volume information, to permit a disk image to be mounted, only requires the two lowest protocol layers. Consequently, the overhead associated with the higher layers can be avoided, thereby enabling the transfer to take place at extremely high speeds.
This capability is particularly advantageous in the case of graphic files, which typically contain large amounts of data. In one implementation of the invention, an image server can be employed to provide large graphic files to remote sites. Referring to
In accordance with another aspect of the invention, disk images can be associated with pre-defined actions to be carried out. For example, it might be desirable to make back-up copies of files stored on a series of floppy disks. Referring to
When the user actuates the script file, all of the identified images are mounted at the user's computer. If desired, a checksum operation can be carried out with respect to each mounted image, to verify its integrity. The specified application is then launched. In this case, the application creates a backup of each imaged disk on a floppy disk at the user's computer. Once the operation has been completed, the image files can be unmounted, if desired, or remain mounted on the user's computer, depending upon the user's preferences.
In this regard, the disk image driver can mark a mounted image as being “owned” by an application or process, such as the disk copy utility. If an imaged volume is owned, the driver checks at regular intervals whether the owning process is still executing. If the process is no longer executing, the volume is marked as unowned and the driver then unmounts it. If no application program is specified in the script file, the identified disk images are simply mounted in response to launching of the script file and marked as unowned. In this case the user must manually unmount the volumes.
The foregoing aspect of the invention can be employed to create a “onebutton” installer that is not limited in the number and/or size of files that can be installed. Referring to
An installation package 96 comprises three elements, namely a conventional installer program 98, the image file tome 94, and a script file 100. The script file identifies the images to be mounted for the installation process, and launches the installer program once the images have been mounted. The program then operates in the normal manner to install the appropriate files from the mounted images onto a computer system. By means of this approach, prior limitations that were placed on installation processes, regarding the number and size of files that can be handled, are overcome, since each image only constitutes a single image file even though it may contain a large number of individual data files. Furthermore, since the user is only required to perform a single action, namely launch the script file, the installation takes place in a true “one-button” manner.
From the foregoing, therefore, it can be seen that the present invention provides an efficient procedure for the dissemination and replication of files in an electronic format which is not limited by the sizes of the files themselves. Through the creation of disk images and mounting of the images at remote computers, ready access is provided to the files in a speedy manner, through the use of low level network communication protocols. Through the availability of different types of formats, backing storage can be used in a manner which is most efficient, taking into account the needs of the user. Furthermore, the mounting and transfer of files is carried out independently of file systems themselves, thereby allowing files to be shared among users of different types of computer system.
It will be appreciated by those of ordinary skill in the art that the present invention can be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The presently disclosed embodiments are therefore considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalence thereof are intended to be embraced therein.