Hybrid replicated shared memory

• US 8,090,926 B2
• Filed: 10/05/2007
• Issued: 01/03/2012
• Est. Priority Date: 10/05/2006
• Status: Active Grant

• Current Assignee: Waratek Limited (Waratek Holdings Ltd.)
• Original Assignee: Waratek Pty Limited
• Inventors: Holt, John M.
• Primary Examiner(s): Thai, Tuan V.
• Application Number:
  US11/973,380
• Publication Number:
  US 20080133871A1
• Time in Patent Office: 1,551 Days
• Field of Search: 711/100, 711/129, 711/153, 711/154, 711/170, 711/173
• US Class Current: 711/173
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

A multiple computer system with hybrid replicated shared memory is disclosed. The local memory (10, 20, . . . 80) of each of the multiple computers M1, M2, . . . Mn is partitioned into a first part (11, 21, . . . 81) and a second part (12, 22, . . . 82). Each of the first parts are identical and each of the second parts are independent. The total memory available to the system is the first memory part plus n times the second memory part, n being the total number of application running multiple computers.

Claims

1. A multiple computer system with replicated shared memory, said system comprising:

a multiplicity of computers each interconnected via a communications network and each executing a different portion of an applications program written to be executed on only a single computer,

wherein each said computer has an independent local memory partitioned into two regions, a first one of said regions having content corresponding to corresponding memory content replicated on at least one other computer, and the second of said regions not corresponding to each other,

wherein changes to the content of replicated memory in the first region of the local memory of any one of said multiple computers are transmitted via said communications network to the corresponding replicated memory in the first region of all the other ones of said multiple computers,

wherein additions and/or deletions to the memory locations in the first region of the local memory of any one of said multiple computers are transmitted via said communications network to the first region of all the other ones of said multiple computers whereby the partition between said first and second regions changes with time.

2. The system of claim 1, wherein said first region occupies less than substantially 50% of the local memory of each said multiple computer.

3. The system as in claim 2, wherein said first region occupies from substantially 5% to substantially 40% of the local memory of each said computer.

4. The system as in claim 3, wherein said first region occupies from substantially 10% to substantially 30% of the local memory of each said computer.

5. A method of partitioning an independent local memory of each computer of a multiple computer system comprising a multiplicity of computers each interconnected via a communications system and each executing a different portion of an applications program written to be executed on only a single computer, said method comprising the step of:

(i) for each said computer partitioning the independent local memory into an allocated application memory, said allocated application memory further partitioned into two regions, a first one of said regions comprising application memory contents replicated in at least on other of said computers, and the second of said regions comprising application memory contents not replicated in any other of said computers,

(ii) transmitting via said communications network changes to the replicated content of memory in the first region of the allocated application memory of any one of said multiple computers to the corresponding replicated memory of at least one other of said multiple computers, and

(iii) changing the partition between said first and second region with time by transmitting via said communications network additions and/or deletions to the memory locations in the first region of the allocated application memory of any one of said multiple computers to the first region of at least one other of said multiple computers.

6. The method of claim 5, further comprising:

(iv) allocating less than substantially 50% of the allocated application memory of each said multiple computer to said first region.

7. The method as in claim 6, including the step of:

(v) allocating from substantially 5% to substantially 40% of said allocated application memory to said first region.

8. The method as in claim 7, including the step of:

(vi) allocating from substantially 10% to substantially 30% of said allocated application memory to said first region.