Digital data storage system with improved data migration

• US 5,276,867 A
• Filed: 12/19/1989
• Issued: 01/04/1994
• Est. Priority Date: 12/19/1989
• Status: Expired due to Term

• Current Assignee: Epoch Systems, Inc. (Dell Technologies Inc.)
• Original Assignee: Epoch Systems, Inc. (Dell Technologies Inc.)
• Inventors: Fortier, Richard, Webber, Neil, Mastors, Robert, Taylor, Tracy, Kenley, Gregory, Pownell, James, Ericson, George, Holland, Chuck, Wallace, John
• Primary Examiner(s): Kulik, Paul V.
• Application Number:
  US07/454,066
• Time in Patent Office: 1,477 Days
• Field of Search: 395/425, 395/600, 395/200, 395/575, 371/10.1
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1402 Saving, restoring, recovering or retrying
• CPC Class Codes: G06F 12/08 in hierarchically structured memory systems, e.g. virtual memory systems
• CPC Class Codes: G06F 12/123 with age lists, e.g. queue, most recently used [MRU] list or least recently used [LRU] list
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

A digital data storage apparatus has primary, secondary and backing storage elements characterized by respectively longer access times. A level detector signals when the quantity of data in the secondary store exceeds a threshold amount. A data migrator responds by moving selected data files from the secondary store to the backing store. The apparatus also includes a baseline back-up element that stores archive copies of a set of selected data files. A full back-up element stores archive copies of those files that, (1) were originally copied to the baseline back-up set but have since changed, or (2) are not otherwise within the baseline back-up set. For those files which were originally copied to the baseline back-up set and which have not changed, the full back-up element stores pointers indicating locations of the respective files in the baseline back-up set.

Claims

1. Digital data storage apparatus having improved backup and providing on-line access to migrated digital files, the apparatus comprising

primary storage means having a first quantity of storage locations and a first access speed, for storing a first plurality of digital files,

secondary storage means having a second quantity of storage locations and a second access speed, for storing a second plurality of digital files, wherein the second quantity of storage locations is greater than the first quantity of storage locations and the second access speed is slower than the first access speed,

backing store means, having a third quantity of storage locations and a third access speed, for storing a third plurality of digital files, wherein the third quantity of storage locations is greater than the second quantity of storage locations and the third access speed is slower than the second access speed, and wherein each of the digital files included in any of the first, second, and third plurality of digital files has an associated characteristic, including file attributes including at least one of, file name, file size, file owner, file protections, time elapsed since a previous access, and type of access,

secondary storage level detection means for detecting how many of the second plurality of storage locations are being used for storing digital files, and for generating a secondary-storage-full signal when more than a preset number of the second plurality of storage locations are being so used,

file migration means, in electrical communication with the secondary storage level detection means, with the secondary storage means, and with the backing store means, for automatically migrating digital files from the secondary storage means to the backing store means at selected times, the file migration means including

file selection means, in electrical communication with the secondary storage means, for selecting, in response to the secondary-storage-full signal, digital files to be migrated from the secondary storage means to the backing store means, and

stage-out means, responsive to the data selection means, for migrating digital files, selected by the data selection means, from the secondary storage means to the backing store means,

a baseline-backup means, coupled to said primary storage means, for retrieving from said primary storage means, digital files having a first associated characteristic, for storing copies thereof, and for generating signals representative of storage locations of respective ones of said copies, wherein said copies are referred to as baseline-backup copies, and

a full-backup, coupled to said primary storage means, for

i. retrieving, from said primary storage means, digital files for which baseline-backup copies are not stored, or for which baseline-backup copies are stored, but which have been updated since a time of their retrieval from said primary storage means for storage as baseline-backup copies, and for

ii. storing, along with the full-backup copies, the file identifiers and the storage location-representative signals associated with files for which baseline-backup copies are stored, but which have not been updated since a time of their retrieval from said primary storage means by said baseline-backup means for storing as baseline-backup copies, wherein said storage location-representative signals are indicative of where the baseline-backup copies are stored.

2. Apparatus according to claim 1, wherein

said file selection means includes file evaluation means, responsive to said file attributes, for selecting files to be migrated from the secondary storage means to the backing store means, and

said stage-out means includes means for moving the selected files to a predetermined staging volume of said backing store means, and for clearing storage locations previously occupied by said selected files in said secondary storage means to make said storage locations available for re-use.

3. Apparatus according to claim 2, further comprising

system catalog means, in electrical communication with at least said secondary storage means, for storing and providing transparent access to catalogued file attribute data representative of said attributes of said files stored in said secondary storage means, and for maintaining said catalogued file attribute data after the data contained in said selected files have been migrated to said backing store means, so that transparent access to file attribute data is provided, independent of whether the file is stored in said secondary storage means or migrated from said secondary storage means.

4. Apparatus according to claim 2 wherein said migration means includes staging trail means for establishing a set of volumes in said backing store means into which selected files from said secondary storage means are migrated,

said staging trail means including means for storing said selected files from said secondary storage means, and for storing file attributes associated with said files, including any of file name, file size, file owner, file protections, time elapsed since a previous access, and type of access,

said staging trail means also including means for storing particular ones of said files in a selected volume contained in said set of volumes.

5. Apparatus according to claim 2, including means for grouping said files into a plurality of file systems, the total number of said file systems being less than the number of said files, and said file selection means including means responsive to selected file attributes for selecting particular files of said file systems to be migrated.

6. Apparatus according to claim 2, including means for grouping said files into a plurality file systems, the total number of said file systems being less than the said number of said files, and

said data selection means including means responsive to a bias value maintained for each of said files contained in said file systems,

said bias value being a function of selected file attributes, each of said attributes given a selected weight, wherein said file attributes include any of, days since last access, file age, file size, number of storage locations contained in said secondary storage means, file residence priority and type of last access,

said selected weight being an adjustable quantity so as to facilitate customization of file migration activity.

7. Apparatus according to claim 2 wherein said stage-out means includes fencepost means for clearing a selected quantity of storage locations previously occupied by said selected files in said secondary storage means, while leaving unmodified a selected fencepost quantity of storage locations in said secondary storage means associated with said selected files, said apparatus allowing a user'"'"'s program to access said selected fencepost quantity without staging in said selected files.

8. Apparatus according to claim 7 wherein the selected fencepost quantity of storage locations is 8192 bytes.

9. Apparatus according to claim 7 including stage-in means for staging in said selected files while said selected fencepost quantity is being accessed.

10. Apparatus according to claim 1,

further comprising file access means, in electrical communication with said primary storage means, said second storage means, and said backing store means, for enabling a user to request user-selected files stored on said backing store means, said file access means including

file stage-in means, in electrical communication with said primary storage means, said second storage means, and said backing store means, for staging-in selected files stored on said backing store means to said primary storage means, and concurrently writing said staged-in files to said secondary storage means for future access, in response to user-asserted access request signals.

11. Apparatus according to claim 10, wherein said stage-in means includes means for maintaining in said backing store means a stored digital image of said staged-in files.

12. Apparatus according to claim 10, wherein said file stage-in means further includes means for selectively staging-in only selected portions of said files, in response to said user-asserted access request signals.

13. Apparatus according to claim 10, wherein said backing store means includes a plurality of physical drives for reading and writing files on a plurality of physically discrete storage volumes resident in said plurality of physical drives, the total number of storage volumes being greater than the number of storage volumes resident in the plurality of physical drives at a given time, and

said file stage-in means including batch stage-in means for executing a batched set of file stage-in operations in response to a set of asserted file names,

said batch stage-in means including means for grouping said file stage-in operations in accordance with storage volumes identified therein.

14. Apparatus according to claim 13, further comprising

drive multiplexor means, in electrical connection with said backing store means, and said stage-in means, for providing transparent access to each volume of said plurality of said storage volumes,

said drive multiplexor means including detection means for detecting whether a storage volume containing a requested file is resident in a particular physical drive and for generating a volume-present signal representative of the presence or absence of the storage volume in a given physical drive.

15. Apparatus according to claim 14, wherein said drive multiplexor means further includes means, responsive to the volume-present signal, for directing file requests to the particular physical drive in which the storage volume containing the requested file is resident.

16. Apparatus according to claim 15, wherein said drive multiplexor means further includes volume insertion means, responsive to a volume request signal, for inserting a storage volume containing a requested file into a selected physical drive.

17. Apparatus according to claim 16, wherein said drive multiplexor means further includes scheduler means, in electrical communication with said volume insertion means, for establishing an ordered schedule of storage volumes to be inserted into a selected physical drive, and for generating said volume request signals in accordance with said schedule, to enable a plurality of storage volumes to time-share a selected physical drive.

18. Apparatus according to claim 17, wherein said volume-request signals include signals representative of relative priority of said volume requests, and

said scheduler means includes priority scheduling means for establishing said schedule in accordance with said volume request priorities, and

preemption means for enabling preemptive insertion of storage volumes corresponding to high priority volume requests into a selected physical drive, in place of storage volumes corresponding to low priority volume requests.

19. Apparatus according to claim 1, wherein said backing store means includes optical storage means for selectively storing and reading digital files in optical form.

20. Apparatus according to claim 19, wherein

said file selection means includes file evaluation means, responsive to said file attributes, for selecting files to be migrated from the secondary storage means to the backing store means, and

said stage-out means includes means for moving the selected files to a predetermined staging volume of said secondary storage means, and, responsive to said movement of selected files to the predetermined staging volume, for clearing storage locations previously occupied by said selected files in said secondary storage means to make said storage locations available for re-use.

21. Apparatus according to claim 20, further comprising

system catalog means, in electrical communication with at least said second storage means, for storing and providing transparent access to catalogued file attribute data representative of said attributes of said files stored in said secondary storage means, and for maintaining said catalogued file attribute data after the data contained in said selected files have been migrated to said backing store means, so that transparent access to file attribute data is provided, independent of whether the file is stored in said secondary storage means or migrated from said secondary storage means.

22. Apparatus according to claim 19,

further comprising file access means, in electrical communication with said primary storage means, said second storage means, and said backing store means, for enabling a user to request user-selected files stored on said backing store means, said file access means including

file stage-in means, in electrical communication with said primary storage means, said second storage means, and said backing store means, for staging-in selected files stored on said backing store means to said primary storage means, and concurrently writing said staged-in files to said secondary storage means for future access, in response to user-asserted access request signals.

23. Apparatus according to claim 22, wherein said stage-in means includes means for maintaining in said backing store means a stored digital image of said staged-in files.

24. Apparatus according to claim 22, wherein said file stage-in means further includes means for selectively staging-in only selected portions of said files, in response to said user-asserted access request signals.

25. Apparatus according to claim 22, wherein said backing store means includes a plurality of physical drives, for reading and writing files on a plurality of physically discrete storage volumes resident in said plurality of physical drives, the total number of storage volumes being greater than the number of storage volumes resident in the plurality of physical drives at a given time, and

said file stage-in means includes means for executing a batched set of file stage-in operations in response to a set of asserted file names, said staged-in files being specified by said file means, thereby enhancing the probability that a user-requested file is resident in one of said plurality of physical drives.

26. Apparatus according to claim 25, further comprising

drive multiplexor means, in electrical connection with said backing store means, and said stage-in means, for providing transparent access to each volume of said plurality of said storage volumes,

said drive multiplexor means including detection means for detecting whether a storage volume containing a requested data file is resident in a particular physical drive and for generating a volume-present signal representative of the presence or absence of the storage volume in a given physical drive.

27. Apparatus according to claim 26, wherein said drive multiplexor means further includes means, responsive to the volume-present signal, for directing file requests to the particular physical drive in which the storage volume containing the requested file is resident.

28. Apparatus according to claim 27, wherein said drive multiplexor means further includes volume insertion means, responsive to a volume request signal, for inserting a storage volume containing a requested file into a selected physical drive.

29. Apparatus according to claim 28, wherein said drive multiplexor means further includes scheduler means, in electrical communication with said volume insertion means, for establishing an ordered schedule of storage volumes to be inserted into a selected physical drive, and for generating said volume request signals in accordance with said schedule, to enable a plurality of storage volumes to time-share a selected physical drive.

30. Apparatus according to claim 29, wherein said volume-request signals include signals representative of relative priority of said volume requests, and

said scheduler means includes priority scheduling means for establishing said schedule in accordance with said volume request priorities, and

preemption means for enabling preemptive insertion of volumes corresponding to high priority volume requests into a selected physical drive, in place of storage volumes corresponding to low priority volume requests.

31. Apparatus according to claim 1, wherein said secondary storage level detection means includes low-water-mark means for generating a low-water-mark signal when less than a low-water-mark level number of the second plurality of storage locations are being used to store digital files, and

said stage-out means includes means coupled to said secondary storage means and responsive to said low-water-mark signal for migrating files from secondary storage means to the backing store means until the secondary storage level detection means generates the low-water-mark signal.

32. Apparatus according to claim 1, wherein said file migration means includes prestage reserve means for migrating, periodically, selected files from said secondary storage means to said backing store means without releasing storage locations on said secondary storage means associated with said selected files,

so that client devices can continue to operate upon said selected files, and so that storage locations in said secondary storage means, occupied by unmodified ones of said selected files may be released immediately in response to a storage utilization level exceeding a selected high-water-mark level.

33. Apparatus according to claim 31 wherein said pre-stage reserve means includes means for migrating files until a selected pre-stage reserve level of secondary storage is attained, said pre-stage reserve level of secondary storage being less than said low-water-mark level.

34. Apparatus according to claim 33 wherein said prestage reserve means includes means, coupled to said stage-out means and responsive to said migrating of files thereby, for initiating said migrating of selected files without releasing storage space.

35. A method for improving backup performance and for providing access to migrated digital files, the method comprising the steps of

configuring a primary storage element to have a first quantity of storage locations and a first access speed, to store a first plurality of digital files,

configuring a secondary storage means to have a second quantity of storage locations, greater than the first quantity of storage locations, and a second access speed, slower than the first access speed, to store a second plurality of digital files,

configuring a backing store element to have a third quantity of storage locations, greater than the second quantity of storage locations, and a third access speed, slower than the second access speed, to store a third plurality of digital files,

configuring each of the digital files, included in the first, second, and third plurality of digital files, to have an associated characteristic, including, file attributes including at least one of, file name, file size, file owner, file protections, time elapsed since a previous access, and type of access,

detecting how many of the second plurality of storage locations are being used for storing digital files,

generating a secondary-storage-full signal when more than a preset number of the second plurality of storage locations are being so used,

automatically migrating digital files from the secondary storage means to the backing store element at selected times, the migration step including the steps of

selecting, in response to the secondary-storage-full signal, digital files to be migrated from the secondary storage means to the backing store element,

staging-out digital files, selected by the selecting step, from the secondary storage element to the backing store means,

retrieving from said primary storage means, digital files having a first associated characteristic storing copies thereof, and generating signals representative of storage locations of respect ones of said copies, wherein said copies are referred to as baseline-backup copies,

retrieving from said primary storage means, digital files for which baseline-backup copies are not stored, or for which baseline-backup copies are not stored, but which have been updated since a time of their retrieval from said primary storage means for storage as baseline-backup copies, and

storing, along with the full-backup copies, the file identifiers and the storage locating-representative signals associated with files for which baseline-backup copies are stored, but which have not been updated since a time of their retrieval from said primary storage means by said baseline-backup means for storing as baseline-backup copies, wherein said storage location-representative signals are indicative of where the baseline-backup copies are stored.