Self-archiving data recording

• US 4,575,827 A
• Filed: 05/18/1984
• Issued: 03/11/1986
• Est. Priority Date: 05/18/1984
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kulakowski, John E.
• Primary Examiner(s): Fears, Terrell W.
• Application Number:
  US06/611,809
• Time in Patent Office: 662 Days
• Field of Search: 365/189, 365/230, 365/51
• US Class Current: 365/230.01
• CPC Class Codes: G06F 16/9017 using directory or table look-up use of a directory or look-up table in file systems G06F16/13

Abstract

Data recording is self-archiving by non-delete recording combined with a directory that grows with each recording in a manner that all previous recordings related to the respective files or data sets are pointed to in the directory. Non-delete recording is applicable to those record media exhibiting indelible characteristics, such as write-once optical media and to rewritable media, such as magnetic recording media of all types. Each directory entry is identified by file or data set name and has one field for each data segment currently a part of the file or data set. Each field has a pointer to its represented data irrespective of when it was recorded. Each directory entry also points to the entry associated with the immediately preceding recording operation on any part of the file or data set such that data changes can be traced using the directory. The directory is stored in one address end portion of the data storing medium while the data is stored in the opposite address end portion--the directory is preferably at the low addresses while the data is at the high addresses. The directory recording is at ever increasing addresses while the data recording is at ever decreasing addresses.

Claims

1. In the method of operating a data storage unit having a continuous address space bounded by first and second addresses and respectively representing data storage areas of said data storage unit addressably accessible by addresses in said address space;

the steps of;

repeatedly receiving records of diverse data for storage in the data storage unit, storing said received data in said data storage areas beginning at said first addressed data storage area and proceeding sequentially at successively addressed ones of said data storage areas toward said second addressed data storage area;

each time any of said received records of data are stored in said data storage areas, creating a record directory entry having an identification of the received and just stored records of data and an address pointer identifying the data storage area in which said received and just stored records of data are stored; and

storing said created record directory entries respectively in data storage areas of said data storage unit beginning at said second address and storing successively created ones of said entries respectively at successive addressed ones of said data storage areas beginning at said second address and proceeding sequentially toward said first address.

2. In the method set forth in claim 1, further including the steps of:

identifying groups of said records of data as data sets;

creating a one of said record directory entries for each of said data sets with each of said created directory entries including address data for all of said records in said identified groups, respectively;

changing a one of said records in a one of said identified groups; and

creating a non-delete record directory entry for the respective data set including address data of said one record directory entry for said one identified group of all records not changed and address data for said changed one of said records and deleting the address data of said one record directory entry for said one identified group of said record that was changed and not deleting the the last mentioned record from said data storage unit nor said one record directory entry for said one identified group.

3. In the method set forth in claim 2, further including the steps of:

receiving a fist plurality of changes to a corresponding first plurality of records stored in said data storage unit and storing said received changed records as set forth in claim 1 without deleting any recorded records of data in the data storage unit; and

creating a second plurality of said non-delete record entries using the ultimate step of claim 2 for changes to records in a corresponding second plurality of said identified groups corresponding to said plurality of changes.

4. In the method set forth in claim 3, further including the step of:

in each of said non-delete record directory entries storing an address pointer indicating the storage address of a record directory entry for the corresponding data set having address data for the records being changed.

5. In the method set forth in claim 4 wherein the method is performed in a data storage unit having a rotatable record storage disk with a data receiving surface means with a plurality of record track shaped and addressable data storage areas in the data receiving surface means and having a radially outwardmost and radially inwardmost record portions, including the steps of:

identifying a one of said track shaped data storage areas in each of said record portions respectively by said first and second addresses;

comparing the address used for last storing a record of data with the address used for last storing a created non-delete record directory entry, when said comparison indicates a predetermined closeness of said compared addresses, indicating that the data receiving surface means is filled, otherwise continue storing any received records of data.

6. In the method set forth in claim 5 further including the step of:

when indicating said data receiving surface means is filled, storing in said data receiving surface means an index to said created non-delete record directory entries having the identifications of said groups associated with the address of the last created ones of the record directory entries, respectively.

7. In the method set forth in claim 5 further including the step of selecting said data receiving surface means to have write once data storage characteristics.

8. In the method of operating a data processing apparatus having a data storage unit having a plurality of addressable data storage areas;

the steps of;

repeatedly receiving records of diverse data for storage in the data storage unit, each of said received records of data having an identifying name therewith, a plurality of received records sharing a common identifying name, some of said received records are being received to update predetermined ones of records currently stored in the data storage unit, respectively, while others of said currently stored records are not being updated;

creating a record directory entry for all of said received records having a common identifying name and which were concurrently received, including in each said created record directory entry said common name and the addresses of the data storage area of all records stored in said data storage unit having said common name, which are not being updated by said concurrently received records, irrespective of time of receipt and storage of such included records and including said currently received records; and

storing each said created record directory entry in a data storage area of said data storage unit and retaining in said data storage unit all current copies of the directory entries and of concurrently stored data.

9. In the method set forth in claim 8, further including the step of:

in each of said created record directory entries storing an address pointer indicating the data unit storage address of a record directory entry for the corresponding common name having address data for the records being updated.

10. In the method set forth in claim 8 wherein the method is performed in a data storage unit having a rotatable record storage disk with a data receiving surface means having a plurality of record track shaped and addressable data storage areas in the data receiving surface means and having radially outwardmost and radially inwardmost record portions, including the steps of:

identifying a one of said record track shaped data storage areas in each of said record portions respectively by first and second addresses of said data storage unit for defining the extremities of said data receiving surface;

storing said data on said data receiving surface beginning at an addressed data storage area identified by said first address and storing ensuing received records of data at addressed data storage areas identified by addresses numerically approaching said second address; and

storing said directory entries at addressed data storage areas beginning with said second address and storing successively created record directory entries in data storages areas having a sequence of addresses proceeding numerically from said second address toward said first address.

11. In the method set forth in claim 10, further including the steps of:

comparing the address last used for storing said received data with the address last used for storing a created record directory entry; and

when said comparison indicates a predetermined numerical closeness of said compared addresses, indicating that the data receiving surface means is filled, otherwise continuing to store any subsequently received records of data.

12. In the method set forth in claim 11 further including the step of:

when indicating said data receiving surface means is filled, storing in said data receiving surface means an index to said created non-delete record directory entries having the identifications of said groups associated with the address of the last created ones of the record directory entries, respectively.

13. In the method set forth in claim 12 further including the step of selecting said data receiving surface means to have write once data storage characteristics.

14. In a data processing apparatus including data storage means for storing data, in combination:

a data storage unit having a large plurality of data storage areas addressable by addresses in an address space bounded by first and second addresses with the addresses in the address space being sequentially increasing from the first address to the second address;

data receiving means connected to said data storage unit for supplying records of data thereto for recording in predetermined ones of said data storage areas, data set means in said data receiving means for identifying said supplied data by record and data set names and for supplying data processing commands which indicate data processing operations, such as recording supplied data, to be performed;

record directory control means connected to said data receiving means including said data set means for responding to said data set means supplied data set name to identify a stored copy of data in said addressable data storage areas identifiable by said supplied data set name, record directory creating means in said record directory control means for creating record directory entries and connected to said addressable data storage areas for storing said created record directory entries, fetching means in said record directory control means connected to said addressable data storage areas for fetching a record directory entry identified by said data set means supplied data set name, merge means in said record directory control means connected to said fetching means and to said receiving means and to said record directory creating means for transferring address pointers from said fetched record directory entry of all records represented by said receiving means supplied data set name that do not correspond to any of said supplied records of data to said record directory creating means such that said record directory creating means creates a record directory entry containing addresses of all records stored in said data storage unit related to said supplied data set name whether supplied by the receiving means or currently stored in the data storage unit and not corresponding to supplied records which update records currently stored in the data storage unit and said record directory creating means storing the just created record directory entry in a unused one of said addressable data storage areas and maintaining a copy of said fetched record directory entry in its data storage area; and

storage control means connected to said data receiving means and to said data storage areas for storing supplied records only in unused data storage areas while maintaining storage of all previous copies of such supplied records whereby the data storage unit stores supplied records and creates record directory entries without deleting any previously stored record directory entries or records of data.

15. In the data processing apparatus set forth in claim 14 further including in combination:

a microprocessor connected to said data receiving means connected to said data set means for receiving said data processing commands and connected to said data storage unit for supplying device commands thereto for causing said data storage unit to receive and supply data and to perform other device operations;

a control store connected to said microprocessor for storing programming indicia to enable said microprocessor to perform control operations to cause the data processing apparatus to function as a data storage subsystem;

first program indicia in said programming indicia for enabling said microprocessor to be said record directory control means when executing said first programming indicia; and

second program indicia in said program indicia for enabling said microprocessor to be said storage control means when executing said second program indicia.

16. In the data processing system set forth in claim 15 wherein said storage control means as embodied by said microprocessor and said second indicia further includes reverse pointer indicia for enabling said microprocessor to store all of said supplied records of data in said data storage unit addressable data storage areas beginning at the addressable data storage area denominated by said first address and then ensuing supplied records of data at successively addressed ones of said addressable data storage areas numerically proceeding from said first toward said second address and to store all of the created record directory entries in said data storage unit beginning at the addressable data storage area denominated by said second address and ensuing ones of said created record directory entries in successive ones of said addressable data storage areas respectively denominated by a sequence of said addresses proceeding from said second address toward said first address, and address compare indicia in said second indicia for enabling said microprocessor to compare the addresses last used to store a record of data and a created record directory entry for indicating that the data storage unit is filled whenever the comparison indicates a predetermined closeness of said two last addresses such that the microprocessor then stops storing any ensuing supplied records of data in said data storage unit.

17. In the data processing system set forth in claim 16 wherein said data storage unit is a disk recorder having a rotatable record storage disk with a data receiving surface having an inner radius at which an inner given one of said data storage areas is located and is addressable by a one of said bounded addresses and having a outer radius at which an outer given one of said data storage areas is located and is addressable by another one of said bounded addresses.

18. In the data processing system set forth in claim 17 wherein said rotatable record storage disk is removably mounted in said data storage unit, said data receiving surface exhibits write-once data recording characteristics, and said data storage unit has optical data recording and sensing apparatus.

19. In the data processing system set forth in claim 18 wherein said data receiving means includes a random access data buffer with a plurality of cache storage areas for respectively temporarily storing records of data to be supplied to said data storage unit and having a cache directory identifying data stored in the data buffer using said supplied record and data set names;

and

a working store connected to said microprocessor for temporarily storing said fetched record directory entries and said created record directory entries.

20. In the data processing system set forth in claim 19 further including a directory index in said working store, said directory index including a list of said data set names for data stored in said data storage unit, the address of the data storage area storing the record directory entry last created for the data represented by the respective data set name being stored with each of said index data set names, respectively.

21. For a data storage unit having a data storage area with a given continuous address space identifiable from area zero to area end, a directory for identifying the locations of data stored in said data storage unit by using said continuous address space, including, in combination:

a first entry stored in said data storage unit in a first data storage area of said data storage unit and containing an identification of a group of related records of data stored in the data storage unit and a plurality of address pointers of said continuous address space pointing respectively to data stored in the data storage unit at diverse ones of said data storage areas and the extent of data storage space storing said first entry being a first extent determined by a fixed amount of data storage space plus a second extent determined by the number of pointers to the stored data; and

a second entry stored in said data storage unit in a second data storage area of said data storage unit and spaced from said first data storage area so that the addresses of said first and second data storage areas are not successive addresses in said continuous address space, said second entry having said identification of data stored and a plurality of address pointers of said continuous address space pointing to stored data stored in said data storage unit as indicated in said first entry by predetermined ones of its said data address pointers, said predetermined ones being less than the number of data address pointers of said first entry and including additional data address pointers to data stored in said data storage unit since said first entry was stored in said data storage unit except that some of said additional address pointers point to stored data as an update of data currently stored in said data storage unit.

22. The invention set forth in claim 21, further including in combination:

said first and second entry being an entry pair;

a plurality of said entry pairs respectively identifying and having address pointers to a like plurality of said groups of related stored records of data; and

all of said first and second entries being stored in respective ones of said data storage areas that are addressable in a sequence of addresses determined by the time of storage of the respective entries and independently of the informational content of any of said entries.

23. The invention set forth in claim 22, further including in combination:

all of said entry pairs being stored in successively addressed ones of said data storage areas beginning at a first address and all of said stored records of data being stored in successively addressed ones of said data storage areas beginning at a second address which is different than said first address.

24. The invention set forth in claim 23, further including in combination:

a record storage disk having a data storing surface means with a inner given data storage area located at an inner radial extremity of the data storing surface means and an outer given data storage area located at an outer radial extremity of the data storing surface means; and

a first one of said given data storage areas being addressable by said first address and a second one of said given data storage areas being addressable by said second address.

25. The invention set forth in claim 24 wherein said record storage disk data storing surface means exhibits write-once data storage characteristics.