Computer architecture and method of operation for multi-computer distributed processing with finalization of objects

• US 8,028,299 B2
• Filed: 10/25/2005
• Issued: 09/27/2011
• Est. Priority Date: 04/21/2005
• Status: Active Grant

• Current Assignee: Waratek Limited (Waratek Holdings Ltd.)
• Original Assignee: Waratek Pty Limited
• Inventors: Holt, John Matthew
• Primary Examiner(s): Ho; Andy
• Assistant Examiner(s): DAO, TUAN C.
• Application Number:
  US11/259,762
• Publication Number:
  US 20060265705A1
• Time in Patent Office: 2,163 Days
• Field of Search: 710/244, 707/206, 707/200, 711/119, 711/135, 717/148, 717/162
• US Class Current: 719/320
• 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 particular, the finalization of JAVA language classes and objects is disclosed (162, 163) so finalization only occurs when the last class or object present on all machines is no longer required.

Claims

1. A multiple computer system having at least one application program each written to operate only on a single computer but running simultaneously on a plurality of computers interconnected by a communications network,

wherein different portions of each of said at least one application program(s) executes simultaneously on different ones of said computers and, for each said portion, a like plurality of identical objects are created, each in the corresponding computer and each having a identical name,

wherein each said computer has an independent local memory, all read requests by each said computer are satisfied by reading only from the corresponding local memory of that computer and from no other computer,

wherein a local memory capacity allocated to each said at least one application program is identical, the total memory capacity available to said each said at least one application program is said allocated memory capacity, and

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

2. The system as claimed in claim 1 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.

3. The system as claimed in claim 2 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.

4. The system as claimed in claim 3 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 need to refer to a corresponding object.

5. The system as claimed in claim 4 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.

6. The system as claimed in claim 2 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.

51. A multiple computer system as in claim 1, wherein:

no local memory location in any of said single computers is accessible for read operations or write operations by any other one of said single computers, and the contents of the local memory are updated by a separate distributed run time means that communicates the updated contents over the communication network; and

the communication network comprises the Internet communication network.

7. A plurality of computers interconnected via a communications link and operating simultaneously at least one application program each written to operate only on a single computer,

wherein each said computer simultaneously executes a different portion of said at least one application program(s), each said computer in operating its application program portion needs, or no longer needs to refer to an object only in local memory physically located in each said computer,

the contents of the local memory utilized by each said computer is fundamentally similar but not, at each instant, identical, wherein each said computer has an independent local memory, all read requests by each said computer are satisfied by reading only from the corresponding local memory of that computer and from no other computer,

wherein the local memory capacity allocated to said at least one application program is identical, the total memory capacity available to said at least one application program is said allocated memory capacity, and

wherein every one of said computers has a finalization routine which deletes a non-referenced object only if each one of said plurality of computers no longer needs to refer to their corresponding object.

8. The plurality of computers as claimed in claim 7 wherein all said distribution update means communicate via said communications link at a data transfer rate which is less than the local memory read rate.

9. The plurality of computers as claimed in claim 7 wherein at least some of said computers are manufactured by different manufacturers or have different operating systems.

10. A method of running simultaneously on a plurality of computers at least one application program each written to operate only on a single computer, each said computer having an independent local memory, said computers being interconnected by means of a communications network, said method comprising:

(i) executing different portions of said application program(s) on different ones of said computers and for each said different portion creating a like plurality of identical objects each in the corresponding computer and each having a identical name, satisfying all read requests by each said computer in executing said different portion by reading only from the corresponding local memory of that computer executing said different portion and from no other computer, wherein the local memory capacity allocated to the or each said application program is identical among said plurality of computers, the total memory capacity available to the or each said application program is said local allocated memory capacity; and

(ii) deleting all said identical objects collectively only when all of said plurality of computers no longer need to refer to their corresponding object.

11. A method as claimed in claim 10 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.

12. A method as claimed in claim 11 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.

13. A method as claimed in claim 12 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.

14. A method as claimed in claim 13 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.

36. A computer program product comprising a set of program instructions stored in a storage medium and operable to permit a plurality of computers to carry out the method as claimed in claim 10.

38. A plurality of computers interconnected via a communication network and operable to ensure consistent initialization of an application program written to operate only on a single computer but running simultaneously on said computers, said computers being programmed to carry out the method as claimed in claim 10.

40. A plurality of computers interconnected via a communication network and operable to ensure consistent initialization of an application program written to operate only on a single computer but running simultaneously on said computers, said computers being loaded with the computer program product as claimed in claim 36.

15. A computer implemented method of operating a plurality of computers to ensure consistent finalization of an application program written to operate only on a single computer but different portions of which are to be executed simultaneously each on a different one of a plurality of computers interconnected via a communications network, each said one of said plurality of computers having at least one local processor and an independent local memory coupled with said processor, said method comprising:

(i) scrutinizing said application program at, or prior to, or after loading to detect each program step defining a finalization routine,

modifying said application program so that all read requests by each said computer in executing said different portion are satisfied by reading only from the corresponding local memory of that computer executing said different portion and from no other computer, the local memory capacity allocated to said application program being identical among said plurality of computers, and the total memory capacity available to said application program is said local allocated memory capacity; and

(ii) modifying said finalization routine to ensure collective deletion of corresponding objects in all said computers only when each one of said computers no longer needs to refer to their corresponding object.

16. The method as claimed in claim 15 wherein said finalization routine is modified to execute to clean-up an object once only and on only one of said computers and when all of said computers no longer need to refer to said object.

17. The method claimed in claim 15 wherein step (ii) comprises the steps of:

(iii) loading and executing said finalization routine on one of said computers, (iv) modifying said finalization routine by said one computer, and (v) transferring said modified finalization routine to each of the remaining computers.

18. The method as claimed in claim 17 wherein said modified finalization routine is supplied by said one computer direct to each of said remaining computers.

19. The method as claimed in claim 17 wherein said modified finalization routine is supplied in cascade fashion from said one computer sequentially to each of said remaining computers.

20. The method claimed in claim 15 wherein step (ii) comprises the steps of:

(vi) loading and modifying said finalization routine on one of said computers, (vii) said one computer sending said unmodified finalization routine to each of the remaining computers, and (viii) each of said remaining computers modifying said finalization routine after receipt of same.

21. The method claimed in claim 20 wherein said unmodified finalization routine is supplied by said one computer directly to each of said remaining computers.

22. The method claimed in claim 20 wherein said unmodified finalization routine is supplied in cascade fashion from said one computer sequentially to each of said remaining computers.

23. The method as claimed in claim 15 including the further step of:

(ix) 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.

24. The method as claimed in claim 15 including the further step of:

(x) 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.

25. The method claimed in claim 16 wherein step (ii) comprises the steps of:

(iii) loading and executing said finalization routine on one of said computers, (iv) modifying said finalization routine by said one computer, and (v) transferring said modified finalization routine to each of the remaining computers.

26. The method as claimed in claim 25 wherein said modified finalization routine is supplied by said one computer direct to each of said remaining computers.

27. The method as claimed in claim 25 wherein said modified finalization routine is supplied in cascade fashion from said one computer sequentially to each of said remaining computers.

28. The method claimed in claim 16 wherein step (ii) comprises the steps of:

(vi) loading and modifying said finalization routine on one of said computers, (vii) said one computer sending said unmodified finalization routine to each of the remaining computers, and (viii) each of said remaining computers modifying said finalization routine after receipt of same.

29. The method claimed in claim 28 wherein said unmodified finalization routine is supplied by said one computer directly to each of said remaining computers.

30. The method claimed in claim 28 wherein said unmodified finalization routine is supplied in cascade fashion from said one computer sequentially to each of said remaining computers.

31. The method as claimed in claim 16 including the further step of:

(ix) 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.

32. The method as claimed in claim 16 including the further step of:

(x) 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.

39. A plurality of computers interconnected via a communication network and operable to ensure consistent initialization of an application program written to operate only on a single computer but running simultaneously on said computers, said computers being programmed to carry out the method as claimed in claim 15.

33. A multiple thread computer processing method in which different individual threads or different groups of threads of a single application program written to operate only on a single computer are simultaneously processed each on a corresponding different one of a plurality of computers each having a local memory and a local memory capacity, each of said plurality of single computers being interconnected via a communications link, and in which objects in the local memory physically associated with the computer processing of each thread have corresponding objects in the local memory of each other said computer, the method comprising:

satisfying all read requests by each said computer of said plurality of computers by reading only from the corresponding local memory of that computer which operates to execute a particular different individual threads or particular different group of threads and from no other computer, the local memory capacity allocated to the application program being identical for each computer, and the total memory capacity available to the application program being limited to said allocated memory capacity; and

collectively deleting all said corresponding objects when each one of said plurality of computers no longer needs to refer to their corresponding object.

34. The method of claim 33, wherein an object residing in the memory associated with one said thread and to be deleted has its identity communicated by the computer of said one thread to a shared table accessible by all other said computers.

35. The method of claim 33, wherein an object residing in the memory associated with one said thread and to be deleted has its identity transmitted to the computer associated with another said thread and is transmitted thereby to a shared table accessible by all said other computers.

37. A computer program product stored in a computer readable storage medium and comprising a set of program instructions which, when executed in a computer, are operable to permit a plurality of computers to carry out a method of operating a plurality of computers to ensure consistent finalization of an application program written to operate only on a single computer, wherein different portions of said set of program instructions are to be executed simultaneously each on a different one of a plurality of computers interconnected via a communications network, each of said plurality of computers having an in dependent local memory, said method comprising:

scrutinizing said application program at, or prior to, or after loading to detect each program step defining a finalization routine;

modifying said application program so that all read requests by each said computer in executing said different portion are satisfied by reading only from the corresponding local memory of that computer executing said different portion and from no other computer,

the local memory capacity allocated to said application program being identical among said plurality of computers, and the total memory capacity available to said application program is said local allocated memory capacity; and

modifying said finalization routine to ensure collective deletion of corresponding objects in all said computers only when each one of said computers no longer needs to refer to their corresponding object.

41. A plurality of computers interconnected via a communication network and operable to ensure consistent initialization of an application program written to operate only on a single computer but running simultaneously on said computers, said computers being loaded with the computer program product as claimed in claim 37.

42. A single computer configured for operating with a plurality of single computers in a multiple computer system and having at least one application program written to operate on only a single computer but running simultaneously on the plurality of single computers interconnected by a communications network;

the single computer comprising;

a local independent memory structure defined in a local independent memory of the single computer and configured to provide execution of application program code of the application program including a plurality of code threads that are written with the intent to execute on and reference 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 single computer of said plurality of single computers with the consequence that all read requests are satisfied locally from said independent local memory of said single computer and from no other computer;

the single computer configured for and executing a different portion of said at least one application program than the other computers of the plurality of single computers, and executing its portion simultaneously with the execution of different portions of the application program on the different other ones of said plurality of computers and, for each portion in said single computer, a plurality of objects are created in its independent local memory while a like plurality of identical objects are created in the independent local memory of the other computers, each object having a identical name; and

means for collectively deleting all said identical objects on said single computer and on the other plurality of computers only when said single computer and each other one of said plurality of computers no longer needs to refer to their corresponding object.

43. The system as claimed in claim 42, wherein the communications network comprises the Internet.

44. The system as claimed in claim 42, wherein each said single computer includes a distributed run time means configured for communication with distributed run time means of other of the plurality of computers and with the distributed run time means of each said computer able to communicate with all other computers so that if a different portion of said application program running on one of said plurality of single computers creates an object in that single computer then the created object is propagated by the distributed run time means of said one computer to all the other computers via their own corresponding run time means.

45. A single computer configured for use with a plurality of different networked single computers that are interconnected via a communications link, the single computer and the plurality of different computers operating simultaneously to execute an application program written to operate on only a single computer, the application program having application program code including a plurality of code threads all intended to execute on and reference a single computer having a single processing unit or symmetric multiple processing units and a single independent local memory with a local memory capacity that is not shared with any other single computer of said plurality of single computers with the consequence that all read requests are satisfied locally from said independent local memory of said single computer and from no other computer;

said single computer simultaneously executes a first portion of said application program and other of said plurality of different networked single computers simultaneously executes a second and other different portion with said first portion;

said single computer, in operating said application program first portion, utilizes a named object only by using a local replica of the named object stored in independent local memory physically located in said single computer with a local memory capacity that is not shared with or accessible by any other of the plurality of different networked single computers;

the contents of the independent local memory utilized by said single computer and by each said plurality of different networked single computers is fundamentally similar but not, at each instant, identical; and

said single computer having and executing a finalization routine which deletes a non-referenced object only if said single computer and each one of said plurality of different networked single computers no longer needs to refer to their corresponding object.

46. A method of running an application program simultaneously on a single computer and simultaneously on a plurality of other single computers interconnected via a communications network, the application program written to operate only on a single computer and having application program code including a plurality of code threads all intended to execute on and reference a single computer having a single processing unit or symmetric multiple processing units and a single independent local memory with a local memory capacity that is not shared with any other single computer of said plurality of other single computers, said method comprising:

(i) simultaneously executing a first portion of the application program on the single computer and second and other different portions of said application program on different ones of said plurality of single computers and for each said first portion and second and other different portions creating a like plurality of identical objects in the single independent local memory of the single computer and in the corresponding single independent local memory of the other single computers, wherein the local memory capacity of the single computer is not shared with or accessible by any other single computer of said plurality of single computers such that all read requests are satisfied locally and not from any other computer, each object having a identical name; and

(ii) deleting all said identical objects in the local memory of the single computer and collectively in the local memories of the other plurality of single computers only when said single computer and all of said plurality of other computers no longer need to refer to their corresponding object in their respective independent local memories.

47. In a single computer, a method of ensuring consistent finalization of an application program written to operate only on one single computer but different portions of which are to be executed simultaneously on the single computer and on each different one of a plurality of computers interconnected with each other and with the single computer via a communications network, the application program having application program code including a plurality of code threads all intended to execute on and reference only one computer having a single processing unit or symmetric multiple processing units and only one independent local memory with a local memory capacity that is not shared with any other computer, said method comprising:

(i) scrutinizing said application program at, or prior to, or after loading on said single computer to detect each application program step defining a finalization routine instruction finalizing an object utilizing said single computer or one of said plurality of other computers in the application program, wherein for each said different portion of the application program a like plurality of identical objects are created in each single independent local memory of the corresponding computer including in the single independent memory of the single computer and with a local memory capacity that is not shared with or accessible by any other computer of said plurality of computers such that all read requests are satisfied locally and not from any other computer and each object having a identical name; and

(ii) modifying said finalization routine to ensure collective deletion of corresponding objects in said single computer and in all said other plurality of computers only when said single computer and each one of said other plurality of computers no longer needs to refer to their corresponding object.

48. In a single computer wherein a first portion of a multi-portion multiple thread processing computer operation is executing in which different individual threads of different ones of the multiple portions of a single application program referencing objects are simultaneously being processed on said single computer and on each one of a different one of a plurality of other computers interconnected via a communications link;

a method comprising;

defining a memory structure in which said referenced objects are defined in an independent local memory of said single computer physically associated with the processing threads of said first portion and have corresponding objects in the local independent memory of each other of said plurality of computers, the single independent local memory of the single computer having a local memory capacity that is not shared with any other of said plurality of other computers such that all read requests are satisfied locally and not from any other computer; and

deleting all said identical objects in the local memory of the single computer and collectively in the local memories of the other plurality of single computers only when said single computer and all of said plurality of other computers no longer need to refer to their corresponding object in their respective independent local memories.

49. A multiple computer system having at least one application program each written to operate on only a single computer but running simultaneously on a plurality of single computers interconnected by a communications network;

the system comprising;

a local independent memory structure defined for each of the plurality of single computers configured to provide execution of application program code of the application program including a plurality of code threads that are written with the intent to execute on and reference a single computer having a single processing unit or symmetric multiple processing units and a single independent local memory with a local memory capacity that is not shared with any other single computer of said plurality of single computers such that all read requests are satisfied locally and not from any other computer;

means for executing different portions of said at least one application program simultaneously on different ones of said computers and for each portion a like plurality of identical objects are created in each independent local memory of the corresponding single computer and each object having a identical name; and

means for collectively deleting all said identical objects only when each one of said plurality of computers no longer needs to refer to their corresponding object.

50. A method of ensuring consistent finalization of an application program written to operate only on a single computer but different portions of which are to be executed simultaneously each on a different one of a plurality of single computers interconnected via a communications network, the application program having application program code including a plurality of code threads all intended to execute on and reference a single computer having a single processing unit or symmetric multiple processing units and a single independent local memory with a local memory capacity that is not shared with any other single computer of said plurality of single computers, said method comprising:

(i) scrutinizing said application program at, or prior to, or after loading to detect each application program step defining a finalization routine instruction finalizing an object utilizing one of said computers in the application program, wherein for each said different portion of the application program a like plurality of identical objects being created in each single independent local memory of the corresponding computer with a local memory capacity that is not shared with or accessible by any other single computer of said plurality of single computers such that all read requests are satisfied locally and from any other computer and each object having a identical name; and

(ii) modifying said finalization routine to ensure collective deletion of corresponding objects in all said single computers only when each one of said single computers no longer needs to refer to their corresponding object.