Disk drive memory

• US 5,077,736 A
• Filed: 02/13/1990
• Issued: 12/31/1991
• Est. Priority Date: 06/28/1988
• Status: Expired due to Term

• Current Assignee: Storage Technology Corp (Oracle Corporation)
• Original Assignee: Storage Technology Corp (Oracle Corporation)
• Inventors: Bowers, John H., Walsh, Robert, Rudeseal, George A., Dunphy, Robert H. Jr.
• Primary Examiner(s): Beausoliel, Robert W.
• Application Number:
  US07/480,019
• Time in Patent Office: 686 Days
• Field of Search: 371/10.1, 371/10.2, 371/2.2, 371/11.1, 371/40.1, 371/40.4, 371/51.1, 364/200, 364/900
• US Class Current: 714/6.32
• CPC Class Codes: G06F 11/1008 in individual solid state devices G06F11/1004 takes precedence
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1662 the resynchronized component or unit being a persistent storage device re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G11B 20/1833 by adding special lists or symbols to the coded information G11B20/1806, G11B20/1866 take precedence
• CPC Class Codes: G11B 5/012 Recording on, or reproducing or erasing from, magnetic disks G11B17/00, G11B19/00 take precedence
• CPC Class Codes: Y10S 360/903 Physical parameter, e.g. form factor

Abstract

The disk drive memory of the present invention uses a large plurality of small form factor disk drives to implement an inexpensive, high performance, high reliability disk drive memory that emulates the format and capability of large form factor disk drives. The plurality of disk drives are switchably interconnectable to form redundancy groups of N+M parallel connnected disk drives to store data thereon. The N+M disk drives are used to store the N segments of each data word plus M redundancy segments. In addition, a pool of R backup disk drives is maintained to automatically substitute a replacement disk drive for a disk drive in a redundancy group that fails during operation. The number N of data segments in each data redundancy group can be varied throughout the disk drive memory to thereby match the characteristics of the input data or operational parameters within the disk drive memory. Furthermore, a group of U unassigned disk drives can be maintained as a stock of disk drives that can be powered up as needed and assigned to either a redundancy group or to the pool of backup disk drives.

Claims

1. A disk memory system for storing data files for associated data processing devices comprising:

a plurality of disk drives;

means for assigning a subset of said plurality of disk drives into at least two redundancy groups, each redundancy group consisting of at least two disk drives;

means responsive to a data file being transmitted to said disk memory system from said associated data processing devices for selecting one of said redundancy groups to store said transmitted data file thereon;

means for writing said transmitted data file and redundancy data associated with said transmitted data file in segments across said at least two disk drives in said selected redundancy group;

means for reserving at least one of said plurality of disk drives as backup disk drives, which backup disk drives are shared in common by said redundancy groups; and

means for identifying one of said at least two disk drives in said subset that fails to function; and

means for switchably connecting one of said backup disk drives in place of said identified failed disk drive.

2. The system of claim 1 further including:

means for reconstructing the segment of said data file written on said identified failed disk drive, using said associated redundancy data.

3. The system of claim 2 further including:

means for writing said reconstructed segment of said data file on to said one backup disk drive.

4. The system of claim 2 wherein said reconstructing means includes:

means for generating said segment written on said identified failed disk drive using said associated redundancy data and said segments of said data file written on said disk drives in said redundancy group.

5. The system of claim 1 wherein said writing means includes:

means for dividing said data file into at least two segments; and

means for generating redundancy data for said segmented data file.

6. The system of claim 5 wherein said writing means further includes:

means for writing each of said segments of said transmitted data file and said segments of said redundancy data on to a different one of said at leas two disk drives in said selected redundancy group.

7. The apparatus of claim 5 wherein said generating means includes:

means for generating multi-bit error codes to produce m redundancy segments for each n segments of said transmitted data file.

8. The apparatus of claim 7 wherein said generating means further includes:

means for generating at least one redundancy segment for said n segments of said transmitted data file; and

means for producing at least one redundancy segment for said n segments of said transmitted data file combined with said generated at least one redundancy segment.

9. The apparatus of claim 7 where one of said redundancy segments is a parity segment.

10. The apparatus of claim 7 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

11. The system of claim 1 further including:

means for maintaining data indicative of said at least two disk drives in said selected redundancy group on which said segments of said transmitted data file and said redundancy data are stored.

12. The system of claim 1 further including:

means responsive to a request for said data file from one of said associated data processing devices for reconstructing said segments of said data file.

13. The system of claim 12 further including:

means responsive to said reconstructing means for transmitting said said reconstructed data file to said one data processing device.

14. The apparatus of claim 1 wherein said assigning means includes:

means for selecting at least two different redundancy group sizes; and

means for storing data indicative of the size of each of said redundancy groups.

15. The apparatus of claim 1 wherein said apparatus further includes:

means for maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where us is a positive integer.

16. The apparatus of claim 15 wherein said maintaining means includes:

means for powering off said u unassigned disk drives.

17. A method of storing data files for data processing devices on an associated disk memory system that includes a plurality of disk drives comprising the steps of:

assigning a subset of said plurality of disk drives into at least two redundancy groups, each redundancy group consisting of at least two disk drives;

selecting, in response to one of said processing devices transmitting a data file to said disk memory system, one of said redundancy groups to store said transmitted data file thereon;

writing said transmitted data file and redundancy data associated with said transmitted data file across said at least two disk drives of said selected redundancy group;

reserving at least one of said plurality of disk drives as backup disk drives, said backup disk drives shared in common by said redundancy groups;

identifying one of said disk drives in said subset that fails to function; and

switchably connecting one of said backup disk drives in place of said identified failed disk drive.

18. The method of claim 17 further including the step of:

reconstructing a segment of said data file written on said identified failed disk drive, using said associated redundancy data.

19. The method of claim 18 further including the step of:

writing said reconstructed segment of said data file on to said one backup disk drive.

20. The method of claim 18 wherein said step of reconstructing includes the steps of:

generating said segment written on said identified failed disk drive using said associated redundancy data and segments of said data file written on said disk drives in said redundancy group.

21. The method of claim 18 wherein said step of writing includes the steps of:

dividing said data file into at least two segments; and

generating redundancy data for said segmented data file.

22. The method of claim 21 wherein said step of writing further includes the step of:

writing each of said segments of said transmitted data file and said segments of said redundancy data on to a different one of said at least two disk drives in said selected redundancy group.

23. The method of claim 21 wherein said step of generating redundancy data includes the step of:

generating multi-bit error codes to produce m redundancy segments for each n segments of said transmitted data file.

24. The method of claim 23 wherein said step of generating multi-bit error codes includes:

generating at least one redundancy segment for said n segments of said transmitted data file; and

producing at least one redundancy segment for said n segments of said transmitted data file combined with said generated at least one redundancy segment.

25. The method of claim 23 where one of said redundancy segments is a parity segment.

26. The method of claim 23 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

27. The method of claim 17 further including the step of:

maintaining data indicative of said at least two disk drives in said selected redundancy group on which said segments of said transmitted data file and said redundancy data are stored.

28. The method of claim 17 further including the step of:

concatenating, in response to a request for aid data file from one of said data processing devices, said segments of said data file.

29. The method of claim 28 further including the step of:

transmitting said concatenated segments of said data file to said one data processing device.

30. The method of claim 17 wherein said step of assigning includes the step of:

selecting at least two different redundancy group sizes; and

storing data indicative of the size of each of said redundancy groups.

31. The method of claim 17 wherein said method further includes the step of:

maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where u is a positive integer.

32. The method of claim 31 wherein said step of maintaining includes the step of:

powering off said u unassigned disk drives.

47. The apparatus of claim 21 wherein said apparatus further includes:

means for maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where u is a positive integer.

48. The apparatus of claim 47 wherein said maintaining means includes:

means for powering off said u unassigned disk drives.

33. A disk memory system for storing data files that are accessible by associated data processing devices comprising:

a plurality of disk drives for storing data thereon;

means for transferring data between said disk memory system and said associated data processing devices;

means for assigning a subset of said disk drives into at least two redundancy groups, each said redundancy group containing n+m disk drives, where n is a positive integer;

means for segmenting each data file received from said associated data processing devices via said transferring means into n segments of data, where n is a positive integer greater than 1;

means responsive to said segmenting means for generating m segments of data redundancy information for said segmented data file, where m is a positive integer greater than 1;

means for selecting one of said redundancy groups to store said received data file and said generated redundancy information thereon;

means for switchably interconnecting said n+m disk drives in said selected redundancy group with said segmenting means to write said n segments of data plus said m segments of redundancy data on to said n+m disk drives of said selected redundancy group;

means for reserving r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by said redundancy groups;

means for identifying one of said n+m disk drives that fails to function;

means for switchably connecting one of said r backup disk drives in place of said identified failed disk drive;

means for reconstructing the segment of said data file written on said identified failed disk drive, using said associated redundancy data; and

means for writing said reconstructed segment of said data file on to said one backup disk drive.

34. In a disk memory system including a plurality of disk drives a method of storing data files that are accessible by associated data processing devices comprising the steps of:

transferring data between said disk memory system and said associated data processing devices;

segmenting each data file received from said associated data processing devices via said transferring means into n segments of data, where n is a positive integer greater than 1;

generating m segments of data redundancy information for said segmented data file, where m is a positive integer greater than 1;

assigning a subset of said plurality of disk drives into at least two redundancy groups, each said redundancy group containing n+m disk drives;

switchably interconnecting said n+m disk drives in one of said redundancy groups with said segmenting means to write said n segments of said received data file plus said m segments of redundancy data on to said n+m disk drives of said one redundancy group;

reserving r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by said redundancy groups;

identifying the one of said disk drives in said subset that fails to function;

switchably connecting one of said backup disk drives in place of said identified failed disk drive;

reconstructing the segment of said data file written on said identified failed disk drive, using said associated redundancy data; and

writing said reconstructed segment of said data file on to said one backup disk drive.

35. A disk memory system for emulating a large form factor disk drive to store data files that are accessible by associated data processing devices comprising:

a plurality of small form factor disk drives for storing data thereon;

means for assigning a subset of said disk drives into at least two redundancy groups, each said redundancy group including n+m of said plurality of disk drives, where n and m are both positive integers greater than 1;

means for segmenting each data file received from said associated data processing devices into n segments of data;

means responsive to said segmenting means for generating m segments of data redundancy information for said segmented data file; and

means for switchably interconnecting said n+m disk drives of one of said redundancy groups with said segmenting means to write said n segments of said received data file plus said m segments of redundancy data on to said n+m selected disk drives of said selected redundancy group.

36. The system of claim 35 further including:

means for reserving r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by all said redundancy groups;

means for identifying one of said n+m disk drives that fails to function; and

means for switchably connecting one of said backup disk drives in place of said failed disk drive.

37. The system of claim 36 further including:

means for reconstructing the segment of said data file written on said failed disk drive, using said associated redundancy data; and

means for writing said reconstructed segment of said data file on to said one backup disk drive.

38. The system of claim 37 wherein said reconstructing means includes:

means for generating said segment written on said failed disk drive using said associated redundancy data and said segments of said data file written on said disk drives in said redundancy group.

39. The system of claim 35 further including:

means for maintaining data indicative of said n+m disk drives in said one redundancy group.

40. The system of claim 35 further including:

means responsive to a request for said data file from one of said associated data processing devices for concatenating said segments of said data file.

41. The system of claim 40 further including:

means responsive to said concatenating means for transmitting said concatenated segments of said data file to said one data processing device.

42. The apparatus of claim 35 wherein said generating means includes:

means for generating multi-bit error codes to produce m redundancy segments for each n segments of said received data file.

43. The apparatus of claim 42 wherein said generating means further includes:

means for generating at least one redundancy segment for said n segments of said received data file; and

means for producing at least one redundancy segment for said n segments of said received data file combined with said generated at least one redundancy segment.

44. The apparatus of claim 42 where one of said redundancy segments is a parity segment.

45. The apparatus of claim 44 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

46. The apparatus of claim 35 wherein said assigning means includes:

means for selecting at least two different redundancy group sizes; and

means for storing data indicative of the size of each of said redundancy groups.

49. In a disk memory system a method of emulating a large form factor disk drive using a plurality of small form factor disk drives to store data files that are accessible by associated data processing devices comprising the steps of:

assigning a subset of said disk drives into at least two redundancy groups, each of said redundancy groups containing n+m disk drives, where both n and m are positive integers greater than 1;

selecting one of said redundancy groups;

segmenting each data file received from said associated data processing devices into n segments of data;

generating m segments of data redundancy information for said segmented data file; and

switchably interconnecting said n+m disk drives of one of said redundancy groups with said segmenting means to write said n segments of said received data file plus said m segments of redundancy data on to said n+m disk drives of said one redundancy group.

50. The method of claim 49 further including the steps of:

reserving r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by said redundancy groups;

identifying one of said subset of disk drives that fails to function; and

switchably connecting one of said backup disk drives in place of said failed disk drive.

51. The method of claim 50 further including the steps of:

reconstructing the segment of said data file written on said failed disk drive, using said associated redundancy data; and

writing said reconstructed segment of said data file on to said one backup disk drive.

52. The method of claim 51 wherein said step of reconstructing includes the steps of:

generating said segment written on said failed disk drive using said associated redundancy data and said segments of said data file written on said disk drives in said redundancy group.

53. The method of claim 52 further including the step of:

maintaining data indicative of said n+m disk drives in said one redundancy group.

54. The method of claim 49 further including the step of:

concatenating, in response to a request for said data file from one of said associated data processing devices, said segments of said data file.

55. The method of claim 54 further including the step of:

transmitting said concatenated segments of said data file to said requesting data processing device.

56. The method of claim 49 wherein said step of generating redundancy data includes the step of:

generating multi-bit error codes to produce m redundancy segments for each n segments of said received data file.

57. The method of claim 56 wherein said step of generating multi-bit error codes includes:

generating at least one redundancy segment for said n segments of said received data file; and

producing at least one redundancy segment for said n segments of said received data file combined with said generated at least one redundancy segment.

58. The method of claim 56 where one of said redundancy segments is a parity segment.

59. The method of claim 56 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

60. The method of claim 49 wherein said step of assigning includes the step of:

selecting at least two different redundancy group sizes; and

storing data indicative of the size of each of said redundancy groups.

61. The method of claim 49 wherein said method further including the step of:

maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where u is a positive integer.

62. The method of claim 61 wherein said step of maintaining includes the step of:

powering off said u unassigned disk drives.

63. A disk memory system for storing data files for one or more associated data processing devices comprising:

a plurality of disk drives;

means for assigning a subset of said plurality of said disk drives to at least two redundancy groups, each redundancy group consisting of n+m disk drives, where n and m are both positive integers greater than 1;

means responsive to a data file being transmitted to said disk drive memory from one of said associated data processing devices for segmenting said transmitted data file into n equal segments;

means for generating m redundancy segments using said n segments of said transmitted data file; and

means for writing said n data segments and said m redundancy segments on one of said redundancy groups.

64. The apparatus of claim 63 further including:

means for selecting r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by said redundancy groups.

65. The system of claim 64 further including:

means for identifying one of said disk drives in a redundancy group that fails to function; and

means for reconstructing the segment of said data file written on said failed disk drive, using said associated redundancy data.

66. The system of claim 65 further including:

means for writing said reconstructed segment of said data file on to one of said backup disk drives.

67. The system of claim 66 wherein said reconstructing means includes:

means for generating said segment written on said identified failed disk drive using said associated redundancy data and said segments of said data file written on said disk drives in said redundancy group.

68. The system of claim 63 further including:

means responsive to a request for said data file from one of said associated data processing devices for concatenating said segments of said data file.

69. The apparatus of claim 63 wherein said generating means includes:

means for generating multi-bit error codes to produce m redundancy segments for each n segments of said transmitted data file.

70. The apparatus of claim 69 wherein said generating means further includes:

means for generating at least one redundancy segment for said n segments of said transmitted data file; and

means for producing at least one redundancy segment for said n segments of said transmitted data file combined with said generated at least one redundancy segment.

71. The apparatus of claim 69 where one of said redundancy segments is a parity segment.

72. The apparatus of claim 69 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

73. The apparatus of claim 63 wherein said assigning means includes:

means for selecting at least two different redundancy group sizes; and

means for storing data indicative of the size of each of said redundancy groups.

74. The apparatus of claim 73 wherein said maintaining means includes:

means for powering off said u unassigned disk drives.

75. The apparatus of claim 63 wherein said apparatus further includes:

means for maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where u is a positive integer.

76. In a disk memory system including a plurality of disk drives a method of storing data files for one or more associated data processing devices comprising the steps of:

assigning a subset of said plurality of said disk drives to at least two redundancy groups, each redundancy group consisting of n+m disk drives, where n and m are both positive integers greater than 1;

segmenting, in response to a data file being transmitted to said disk drive memory from one of said associated data processing devices, said transmitted data file into n equal segments.

generating m redundancy segments using said n segments of said transmitted data file; and

writing said n data segments and said m redundancy segments on one of said redundancy groups.

77. The method of claim 76 further including the step of:

selecting r of said plurality of disk drives as backup disk drives, where r is a positive integer, which backup disk drives are shared in common by said redundancy groups.

78. The method of claim 77 further including the step of:

identifying one of said disk drives in a redundancy group that fails to function; and

reconstructing the segment of said data file written on said failed disk drive, using said associated redundancy data.

79. The method of claim 78 further including the step of:

writing said reconstructed segment of said data file on to one of said backup disk drives.

80. The method of claim 79 wherein said step of reconstructing includes the steps of:

generating said segment written on said failed disk drive using said associated redundancy data and said segments of said data file written on said disk drives in said redundancy group.

81. The method of claim 76 further including the step of:

concatenating, in response to a request for said data file from one of said associated data processing devices, said segments of said data file.

82. The method of claim 76 wherein said step of generating redundancy data includes the step of:

generating multi-bit error codes to produce m redundancy segments for each n segments of said transmitted data file.

83. The method of claim 82 wherein said step of generating multi-bit error codes includes:

generating at least one redundancy segment for said n segments of said transmitted data file; and

producing at least one redundancy segment for said n segments of said transmitted data file combined with said generated at least one redundancy segment.

84. The method of claim 82 where one of said redundancy segments is a parity segment.

85. The method of claim 82 wherein at least two of said m redundancy segments are multi-bit error detection and correction codes.

86. The method of claim 76 wherein said step of assigning includes the step of:

selecting at least two different redundancy group sizes; and

storing data indicative of the size of each of said redundancy groups.

87. The method of claim 76 wherein said method further includes the step of:

maintaining u of said plurality of disk drives as unassigned, being neither data, redundancy or backup disk drives, where u is a positive integer.

88. The method of claim 87 wherein said step of maintaining includes the step of:

powering off said u unassigned disk drives.