Multi-computer distributed processing with replicated local memory exclusive read and write and network value update propagation

• US 7,788,314 B2
• Filed: 10/25/2005
• Issued: 08/31/2010
• Est. Priority Date: 04/23/2004
• Status: Active Grant

• Current Assignee: Waratek Limited (Waratek Holdings Ltd.)
• Original Assignee: Waratek Pty Limited
• Inventors: Holt, John Matthew
• Primary Examiner(s): Avellino; Joseph E
• Assistant Examiner(s): Rashid; Harunur
• Application Number:
  US11/259,885
• Publication Number:
  US 20060242464A1
• Time in Patent Office: 1,771 Days
• Field of Search: 709/202, 709/201, 709/203, 712/228, 717/168, 717/169, 717/170, 717/171, 717/172, 707/8, 718/102, 719/316
• US Class Current: 709/201
• CPC Class Codes: G06F 15/16 Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs coordinating program control therefor G06F9/52; in regulating and control system G05B
• CPC Class Codes: G06F 8/456 Parallelism detection

Abstract

The present invention discloses a modified computer architecture (50, 71, 72) which enables an applications program (50) to be run simultaneously on a plurality of computers (M1, . . . Mn). Shared memory at each computer is updated with amendments and/or overwrites so that all memory read requests are satisfied locally. During initial program loading (75), or similar, instructions which result in memory being re-written or manipulated are identified (92). Additional instructions are inserted (103) to cause the equivalent memory locations at all computers to be updated. In addition, the initialization of JAVA language classes and objects is disclosed (162, 163) so all memory locations for all computers are initialized in the same manner. The finalization of JAVA language classes and objects is also disclosed (162, 163) so finalization only occurs when the last class or object present on all machines is no longer required. During initial program loading (75), or similar, instructions which result in the application program (50) acquiring (or releasing) a lock on a particular asset (50A, 50X-50Y) (synchronization) are identified. Additional instructions are inserted (162, 163) to result in a modified synchronization routine with which all computers are updated.

Claims

1. A method of running simultaneously on a plurality of computers an application program written to operate on only a single computer but modified by a modification routine to run simultaneously on the plurality of computers, said computers being interconnected by a communications network, said method comprising the steps of:

defining a local independent memory structure in the local independent memory of each one of the plurality of computers, the plurality of computers configured to provide execution of different portions of the application program that is written for execution on and that references only a single computer having a single processing unit or symmetric multiple processing units and the single independent local memory with a local memory capacity that is not shared with any other computer of said plurality of computers;

replacing the application program with a modified application program on each of said computers before execution of different portions of said application program on different ones of said plurality of computers so that the processing unit or units of each of said plurality of computers only reads from and writes to the local independent memory structure in its own local independent memory;

executing the different portions of said modified application program on different ones of said plurality of computers and for each said different portion creating and storing an initial contents of each of said substantially identically named objects each in the corresponding local independent memory of each of the plurality of computers;

during execution of said different portions;

(a) maintaining and storing a current contents of each said identically named objects in each local independent memory location of each of the plurality of computers;

(b) satisfying all reading of contents by the processor or processors of said computer carrying out the execution which generates the demand to read memory using a current contents of a memory location stored in its own local independent memory; and

(c) satisfying all writing of contents by the processor or processors of said of said computer carrying out the execution which generates the demand to write to memory using a current contents of a memory location stored in its own local independent memory; and

propagating updates of the contents of said substantially identically named objects from all of said computers that write a contents to their own local independent memory to the corresponding local independent memory location of each of the other computers over said communications network and storing the updated contents as the current contents in each said local independent memory.

2. The method as claimed in claim 1, wherein said modified application program includes modification to portions of the application program that manipulate or change values or contents of a memory location by inserting a updating propagation routine that when executed causes memory location values or contents that are manipulated or changed by a portion of the application program executing on one of said computers to be propagated over the communications network to other of the plurality of computers after being written to the local independent memory of the computer carrying out the execution which generates the write to its own local independent memory.

3. The method as claimed in claim 1, wherein said replacing the application program with a modified application program further including performing modification to portions of the application program that manipulate or change values or contents of a memory location by inserting a updating propagation routine that when executed causes memory location values or contents that are manipulated or changed by a portion of the application program executing on one of said computers to be propagated over the communications network to other of the plurality of computers after being written to the local independent memory of the computer carrying out the execution which generates the write to its own local independent memory.

4. The method as claimed in claim 1, wherein the updates are propagated through a distributed runtime executing as a different process from said modified application program on each of said plurality of computers, said distributed runtime execution not delaying execution of said modified application program on any of said plurality of computers.

5. The method as claimed in claim 1, wherein said local independent memory of each of said plurality of computers stores a current value of everyone of said substantially identically named objects except for a propagation update delay time.

6. The method as claimed in claim 1, comprising the further steps of:

creating the initial contents of each of said identically named objects substantially the same;

deleting all said identical objects collectively when all of said plurality of computers no longer need to refer to their corresponding object;

requiring any of said computers wishing to utilize a named object therein to acquire an authorizing lock which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished;

if a portion of said application program running on one of said computers creates an object in that computer, then the created object is propagated to all of the other computers via said communications network; and

if a portion of said application program running on one of said computers changes the contents of an object in that computer, then the change in content of said object is propagated to all of the other computers via said communications network to change the content of the corresponding object in each of said other computers;

modifying said application program before, during or after loading by inserting an initialization routine to modify each instance at which said application program creates an object, said initialization routine propagating every object created by one computer to all said other computers; and

modifying said application program before, during or after loading by inserting an updating propagation routine to modify each instance at which said application program writes to memory, said updating propagation routine propagating every memory write by one computer to all said other computers;

providing each said computer with a distributed run time means to communicate between said computers via said communications network;

providing a shared table accessible by each said distributed run time means and in which is stored the identity of any computer which no longer requires to access an object, together with the identity of the object;

associating a counter means with said shared table, said counter means storing a count of the number of said computers which no longer require to access said object; and

providing an additional computer on which said shared program does not run and which hosts said shared table and counter, said additional computer being connected to said communications network.

7. A computer program product stored on a tangible computer readable media or existing in electronic or digital form anywhere and operable to permit at least one computers to carry out the method as claimed in claim 1.

46. A computer program product comprising a set of program instructions stored in a storage medium or existing in electronic or digital form anywhere and operable to permit at least one computers to carry out the method as claimed in claim 1.

8. A multiple computer system comprising:

a plurality of computers interconnected by a communications network, each said computer having a processor or processors and including an independent local memory whose local memory capacity is not shared with the local independent memories of other of the computers;

the plurality of computers configured to execute simultaneously on the plurality of computers an application program written to operate on only a single computer but modified by a modification routine to execute different portions of said application program simultaneously on the plurality of computers,

a local independent memory structure defined in the local independent memory of each one of the plurality of computers the plurality of computers configured to provide execution of different portions of the application program that is written for execution on and that references only a single computer having a single processing unit or symmetric multiple processing units and the single independent local memory with a local memory capacity that is not shared with any other computer of said plurality of computers;

program modification means for replacing the application program with a modified application program on each of said computers before execution of different portions of said application program on different ones of said plurality of computers so that the processing unit or units of each of said plurality of computers only reads from and writes to the local independent memory structure in its own local independent memory;

means for executing the different portions of said modified application program on different ones of said plurality of computers and for each said different portion creating and storing an initial contents of each of said substantially identically named objects each in the corresponding local independent memory of each of the plurality of computers;

means operable during said execution of said different portions;

(a) for maintaining and storing a current contents of each said identically named objects in each local independent memory location of each of the plurality of computers;

(b) for satisfying all reading of contents by the processor or processors of said computer carrying out the execution which generates the demand to read memory using a current contents of a memory location stored in its own local independent memory; and

(c) for satisfying all writing of contents by the processor or processors of said of said computer carrying out the execution which generates the demand to write to memory using a current contents of a memory location stored in its own local independent memory; and

a distributed run time means configured on each said computer for propagating updates of the contents of said substantially identically named objects from all of said computers that write a contents to their own local independent memory to the corresponding local independent memory location of each of the other computers over said communications network and storing the updated contents as the current contents in each said local independent memory.

9. A method of running simultaneously on a plurality of computers at least one application program each written to operate on only a single computer having processing unit and a single local independent memory coupled with the processing unit that is not shared with the independent local memory of each other of the plurality of computers, the application program modified prior to execution on each of said plurality of computers to execute different portions of the application program simultaneously on the plurality of computers in a consistent and coordinated fashion across all the plurality of computers, said computers being interconnected by means of a communications network, said method comprising the steps of:

(i) executing said different portions of said at least one application program on different ones of said computers and for each said different portion creating a plurality of substantially identical objects each in the local independent memory of each of the plurality of corresponding computers; and

during said execution of said different portions;

(a) storing a value of each said objects in each local independent memory location of each of the plurality of computers; and

(b) satisfying all read and write requests for values of said objects exclusively from the value stored in local independent memory of the computer that executes an application program portion generating the read request or write request.

10. The method as claimed in claim 9, comprising the further step of:

(ii) naming each of said plurality of substantially identical objects with a substantially identical name.

11. The method as claimed in claim 9 wherein:

each said object having a substantially identical name.

12. The method as claimed in claim 11, comprising the further step of:

(iii) creating the initial contents of each of said identically named objects substantially the same.

13. The method as claimed in claim 11, comprising the further step of:

(iii) deleting all said identical objects from the local independent memories of the plurality of computers collectively when all of said plurality of computers no longer need to refer to their corresponding object.

14. The method as claimed in claim 11, comprising the further step of:

(iii) requiring any of said computers wishing to utilize a named object therein to acquire an authorizing lock which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished.

15. The method as claimed in claim 9, wherein each computer having an independent local memory with a local memory capacity that is not shared with any other computer of said plurality of computers.

16. The method as claimed in claim 15, wherein each object accessible only by the corresponding portion of said application program.

17. The method as claimed in claim 10, comprising the further step of:

(ii) naming each of said plurality of substantially identical objects with a substantially identical name.

18. The method as claimed in claim 9, wherein:

each computer having an independent local memory;

each object accessible only by the corresponding portion of said application program;

and the method comprising the further steps of;

(ii) naming each of said plurality of substantially identical objects with a substantially identical name;

(iii) creating the initial contents of each of said identically named objects substantially the same;

(iv) deleting all said identical objects collectively when all of said plurality of computers no longer need to refer to their corresponding object; and

(v) requiring any of said computers wishing to utilize a named object therein to acquire an authorizing lock which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished.

19. The method as claimed in claim 9, comprising the further steps of:

(ii) creating the initial contents of each of said identically named objects substantially the same;

(iii) deleting all said identical objects collectively when all of said plurality of computers no longer need to refer to their corresponding object; and

(iv) requiring any of said computers wishing to utilize a named object therein to acquire an authorizing lock which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished.

20. The method as claimed in claim 19, comprising the further step of:

(v) if a portion of said application program running on one of said computers creates an object in that computer, then the created object is propagated to all of the other computers via said communications network.

21. The method as claimed in claim 20, including the further step of:

(vi) modifying said application program before, during or after loading by inserting an initialization routine to modify each instance at which said application program creates an object, said initialization routine propagating every object created by one computer to all said other computers.

22. A method as claimed in claim 19, including the further step of:

(vii) providing each said computer with a distributed run time means to communicate between said computers via said communications network.

23. A method as claimed in claim 22, including the further step of:

(viii) providing a shared table accessible by each said distributed run time means and in which is stored the identity of any computer which no longer requires to access an object, together with the identity of the object.

24. A method as claimed in claim 23, including the further step of:

(ix) associating a counter means with said shared table, said counter means storing a count of the number of said computers which no longer require to access said object.

25. A method as claimed in claim 24, including the further step of:

(x) providing an additional computer on which said shared program does not run and which hosts said shared table and counter, said additional computer being connected to said communications network.

26. A method as claimed in claim 22, including the further step of:

(xi) providing a shared table accessible by each said distributed run time means and in which is stored the identity of any computer which currently has to access an object, together with the identity of the object.

27. A method as claimed in claim 26, including the further step of:

(xii) associating a counter means with said shared table, said counter means storing a count of the number of said computers which seek access to said object.

28. A method as claimed in claim 27, including the further step of:

(xiii) providing an additional computer on which said shared program does not run and which hosts said shared table and counter, said additional computer being connected to said communications network.

29. The method as claimed in claim 16, comprising the further step of:

(ii) naming each of said plurality of substantially identical objects with a substantially identical name.

30. The method as claimed in claim 29, comprising the further step of:

(iii) if a portion of said application program running on one of said computers changes the contents of an object in that computer, then the change in content of said object is propagated to all of the other computers via said communications network to change the content of the corresponding object in each of said other computers.

31. The method as claimed in claim 30, including the further step of:

(iv) modifying said application program before, during or after loading by inserting an updating propagation routine to modify each instance at which said application program writes to memory, said updating propagation routine propagating every memory write by one computer to all said other computers.

32. The method as claimed in claim 31, including the further step of:

(v) modifying said application program utilizing a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

33. The method as claimed in claim 30, including the further step of:

(vi) transferring the modified application program to all said computers utilizing a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

34. The method as claimed in claim 12, comprising the further step of:

(iii) if a portion of said application program running on one of said computers creates an object in that computer, then the created object is propagated to all of the other computers via said communications network.

35. The method as claimed in claim 34, including the further step of:

(iv) modifying said application program before, during or after loading by inserting an initialization routine to modify each instance at which said application program creates an object, said initialization routine propagating every object created by one computer to all said other computers.

36. The method as claimed in claim 35, including the further step of:

(v) modifying said application program utilizing a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

37. The method as claimed in claim 35, including the further step of:

(vi) transferring the modified application program to all said computers utilizing a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

38. A method as claimed in claim 13, including the further step of:

(iii) providing each said computer with a distributed run time means to communicate between said computers via said communications network.

39. A method as claimed in claim 38, including the further step of:

(iv) providing a shared table accessible by each said distributed run time means and in which is stored the identity of any computer which no longer requires to access an object, together with the identity of the object.

40. A method as claimed in claim 39, including the further step of:

(v) associating a counter means with said shared table, said counter means storing a count of the number of said computers which no longer require to access said object.

41. A method as claimed in claim 40, including the further step of:

(vi) providing an additional computer on which said shared program does not run and which hosts said shared table and counter, said additional computer being connected to said communications network.

42. A method as claimed in claim 14, including the further step of:

(iii) providing each said computer with a distributed run time means to communicate between said computers via said communications network.

43. A method as claimed in claim 42, including the further step of:

(iv) providing a shared table accessible by each said distributed run time means and in which is stored the identity of any computer which currently has to access an object, together with the identity of the object.

44. A method as claimed in claim 43, including the further step of:

(v) associating a counter or counter means with said shared table, said counter means storing a count of the number of said computers which seek access to said object.

45. A method as claimed in claim 44, including the further step of:

(vi) providing an additional computer on which said shared program does not run and which hosts said shared table and counter, said additional computer being connected to said communications network.

91. The method as claimed in claim 9, further comprising:

propagating updates of the value of said objects from all of said computers that write to a value or manipulate a value to their own local independent memory during execution over the communications network to the corresponding local independent memory location of each of the other computers and storing the updated value as the current value.

47. A multiple computer system comprising:

a plurality of computers interconnected by a communications network, each said computer having a processor or processors and including an independent local memory whose local memory capacity is not shared with the local independent memories of other of the computers;

the plurality of computers having at least one application program each written to operate only on a single computer but modified by a modification routine prior to execution on each of said plurality of computers to execute different portions of the application program simultaneously on the plurality of computers, wherein said different portions of said application program(s) execute substantially simultaneously on different ones of said plurality of computers; and

for each different portion a like plurality of substantially identical objects are created, each in the corresponding local independent memory of each one of the plurality of computers; and

means configured during said execution of said different portions;

(a) for storing a value of each said objects in each local independent memory location of each of the plurality of computers; and

(b) for satisfying all read and write requests for values of said objects exclusively from the value stored in local independent memory of the computer that executes an application program portion generating the read request or write request.

48. The system as claimed in claim 47, wherein each of said like plurality of substantially identical objects having a corresponding name or identifier on each of said plurality of computers.

49. The system as claimed in claim 47, wherein each of said like plurality of substantially identical objects having a substantially identical name.

50. The system as claimed in claim 47, wherein each of said plurality of substantially identical objects has a substantially identical name.

51. The system as claimed in claim 47, wherein each computer has an independent local memory accessible only by the corresponding portion of said application program(s).

52. The system as claimed in claim 47, wherein the initial contents of each of said identically named objects is substantially the same.

53. The system as claimed in claim 47, wherein all said identical objects are collectively deleted when each one of said plurality of computers no longer needs to refer to their corresponding object.

54. The system as claimed in claim 47, wherein said system including a lock means applicable to all said computers wherein any computer wishing to utilize a named object therein acquires an authorizing lock from said lock means which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished.

55. The system as claimed in claim 47, wherein said portions are different portions.

56. The system as claimed in claim 52, wherein each of said plurality of substantially identical objects has a substantially identical name.

57. The system as claimed in claim 47, wherein:

each of said like plurality of substantially identical objects having a substantially identical name;

the initial contents of each of said identically named objects is substantially the same;

all said identical objects are collectively deleted when each one of said plurality of computers no longer needs to refer to their corresponding object; and

said system including a lock means applicable to all said computers wherein any computer wishing to utilize a named object therein acquires an authorizing lock from said lock means which permits said utilization and which prevents all the other computers from utilizing their corresponding named object until said authorizing lock is relinquished.

58. The system as claimed in claim 57, wherein said lock means includes an acquire lock routine and a release lock routine, and both said routines are included in modifications made to said application program running on all said computers.

59. The system as claimed in claim 58, wherein said lock means further includes a shared table listing said named objects in use by any said computer, a lock status for each said object, and a queue of any pending lock acquisitions.

60. The system as claimed in claim 59, wherein said lock means is located within an additional computer not running said application program and connected to said communications network.

61. The system as claimed in claim 57, wherein each said computer includes a distributed run time means with the distributed run time means of each said computer able to communicate with all other computers whereby if a portion of said application program(s) running on one of said computers creates an object in that computer then the created object is propagated by the distributed run time means of said one computer to all the other computers.

62. The system as claimed in claim 57, wherein each said computer includes a distributed run time means with the distributed run time means of each said computer able to communicate with all other computers whereby if a portion of said application program(s) running on one of said computers no longer needs to refer to an object in that computer then the identity of the unreferenced object is transmitted by the distributed run time means of said one computer to a shared table accessible by all the other computers.

63. The system as claimed in claim 57, wherein each said application program is modified before, during, or after loading by inserting an initialization routine to modify each instance at which said application program creates an object, said initialization routine propagating every object newly created by one computer to all said other computers.

64. The system as claimed in claim 63, wherein the application program is modified in accordance with a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

65. The system as claimed in claim 63, wherein said modified application program is transferred to all said computers in accordance with a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

66. The system as claimed in claim 57, wherein the local memory capacity allocated to the, or each, said application program is substantially identical and the total memory capacity available to the, or each, said application program is said allocated memory capacity.

67. The system as claimed in claim 66, wherein all said computers include a distribution update means each of which communicates via said communications link at a data transfer rate which is substantially less than the local memory read rate.

68. The system as claimed in claim 67, wherein at least some of said computers are manufactured by different manufacturers and/or have different operating systems.

69. The system as claimed in claim 51, wherein each of said plurality of substantially identical objects has a substantially identical name.

70. The system as claimed in claim 69, wherein each said computer includes a distributed run time means with the distributed run time means of each said computer able to communicate with all other computers whereby if a portion of said application program(s) running on one of said computers changes the contents of an object in that computer then the change in content for said object is propagated by the distributed run time means of said one computer to all other computers to change the content of the corresponding object in each of said other computers.

71. The system as claimed in claim 70, wherein each said application program is modified before, during, or after loading by inserting an updating propagation routine to modify each instance at which said application program writes to memory, said updating propagation routine propagating every memory write by one computer to all said other computers.

72. The system as claimed in claim 71, wherein the application program is modified in accordance with a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

73. The system as claimed in claim 72, wherein said modified application program is transferred to all said computers in accordance with a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

74. The system as claimed in claim 52, wherein each said computer includes a distributed run time means with the distributed run time means of each said computer able to communicate with all other computers whereby if a portion of said application program(s) running on one of said computers creates an object in that computer then the created object is propagated by the distributed run time means of said one computer to all the other computers.

75. The system as claimed in claim 74, wherein each said application program is modified before, during, or after loading by inserting an initialization routine to modify each instance at which said application program creates an object, said initialization routine propagating every object newly created by one computer to all said other computers.

76. The system as claimed in claim 75, wherein said inserted initialization routine modifies a pre-existing initialization routine to enable the pre-existing initialization routine to execute on creation of the first of said like plurality of objects, and to disable the pre-existing initialization routine on creation of all subsequent ones of said like plurality of objects.

77. The system as claimed in claim 76, wherein the application program is modified in accordance with a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

78. The system as claimed in claim 74, wherein said modified application program is transferred to all said computers in accordance with a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

79. The system as claimed in claim 53, wherein each said computer includes a distributed run time means with the distributed run time means of each said computer able to communicate with all other computers whereby if a portion of said application program(s) running on one of said computers no longer needs to refer to an object in that computer then the identity of the unreferenced object is transmitted by the distributed run time means of said one computer to a shared table accessible by all the other computers.

80. The system as claimed in claim 79, wherein each said application program is modified before, during, or after loading by inserting a finalization routine to modify each instance at which said application program no longer needs to refer to an object.

81. The system as claimed in claim 80, wherein said inserted finalization routine modified a pre-existing finalization routine to enable the pre-existing finalization routine to execute if all computers no longer need to refer to their corresponding object, and to disable the pre-existing finalization routine if at least one computer does not to refer to a corresponding object.

82. The system as claimed in claim 81, wherein the application program is modified in accordance with a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

83. The system as claimed in claim 79, wherein said modified application program is transferred to all said computers in accordance with a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

84. The system as claimed in claim 54, wherein said lock means includes an acquire lock routine and a release lock routine, and both said routines are included in modifications made to said application program running on all said computers.

85. The system as claimed in claim 84, wherein said lock means further includes a shared table listing said named objects in use by any said computer, a lock status for each said object, and a queue of any pending lock acquisitions.

86. The system as claimed in claim 85, wherein said lock means is located within an additional computer not running said application program and connected to said communications network.

87. The system as claimed in claim 84, wherein each said application program is modified before, during, or after loading by inserting said acquire lock routine and said release lock routine to modify each instance at which said application program acquires and releases respectively a lock on an object.

88. The system as claimed in claim 85, wherein the application program is modified in accordance with a procedure selected from the group of procedures consisting of re-compilation at loading, pre-compilation prior to loading, compilation prior to loading, just-in-time compilation, and re-compilation after loading and before execution of the relevant portion of application program.

89. The system as claimed in claim 84, wherein said modified application program is transferred to all said computers in accordance with a procedure selected from the group consisting of master/slave transfer, branched transfer and cascaded transfer.

90. The system as claimed in claim 47, further comprising a distributed runtime component adapted to provide an exchange of program and/or data over a communications link between the plurality of computers.