Fault tolerant memory

• US 6,223,301 B1
• Filed: 09/30/1997
• Issued: 04/24/2001
• Est. Priority Date: 09/30/1997
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Compaq Computer Corporation (HP Inc.)
• Inventors: Santeler, Paul A., Olarig, Sompong P., Jansen, Kenneth A.
• Primary Examiner(s): Beausoleil, Jr., Robert W.
• Assistant Examiner(s): ELISCA, PIERRE E
• Application Number:
  US08/940,282
• Time in Patent Office: 1,302 Days
• Field of Search: 714/5, 714/6, 714/7, 714/8, 714/13, 714/710, 714/768, 714/763, 714/764, 714/758, 714/752, 711/114, 711/152, 711/170, 365/233, 365/230, 365/200
• US Class Current: 714/6.2
• CPC Class Codes: G06F 11/1044 with specific ECC/EDC distribution
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD

Abstract

A computer system includes memory modules, a central processing unit and a memory controller. The memory controller is configured to access the memory modules in response to interaction with the central processing unit and define a fault tolerant memory array with the memory modules. Each memory module stores first data represented by second data stored by the other memory modules.

Claims

1. A computer system comprising:

memory chip units each comprising individual memory chip modules removably connected by respective memory chip module connectors to a memory bus coupled to a memory controller;

a central processing unit; and

said memory controller configured to access the memory chip units in response to interaction with the central processing unit and define a fault tolerant memory array with the memory chip units, each memory chip unit storing first data represented by second data stored by the other memory chip units;

said memory controller operable to;

correct corrupted first data retrieved from a memory chip unit using error correction code associated with the retrieved data and when the retrieved first data from said memory chip unit cannot be corrected by said error correction code, restoring said first data stored in said memory chip unit by replacement with said second data.

2. The computer system of claim 1, wherein the first data comprises a mirror image of the second data.

3. The computer system of claim 1, wherein the fault tolerant array comprises a RAID-type array.

4. The computer system of claim 1 wherein the use of the second data to restore includes exclusive ORing together the second data from said other memory chip units.

5. The computer system of claim 1, wherein the first and second data comprise a stripe of data stored in all of the memory chip units.

6. The computer system of claim 5, wherein the stripe has contiguous blocks of data, and the first data comprises a portion of one of the blocks.

7. The computer system of claim 1, wherein the memory chip units comprise single in-line memory modules (SIMMs).

8. The computer system of claim 1, wherein the memory chip units comprise double in-line memory modules (DIMMs).

9. The computer system of claim 1, wherein the memory chip units comprise Rambus in-line memory modules (RIMMs).

10. The computer system of claim 1, wherein the slot connectors comprise hot plug slot connectors.

11. A computer system comprising:

at least one processor coupled to a disk storage device and to a fault tolerant redundant memory chip array comprising a plurality of memory chip modules removably connected to a memory bus, said memory chip modules storing stripes of data and parity data associated with respective ones of said stripes of data;

said processor and said redundant memory array coupled to a memory controller operable under control of said processor to write data to and to retrieve data from said redundant memory array;

said memory controller also operable to;

detect errors in corrupted data stored by said memory chip modules in said stripes of data, generate error correction code information associated with the corrupted data, and store the error correction code information in the memory chip module which stores the corrupted data; and

determine whether corrupted data retrieved from a memory module can be corrected using the stored error correction code information associated with that corrupted data;

if so, correcting the corrupted data using the associated error correction code information;

if not, restoring the corrupted data stored by said memory module by replacement with data derived from uncorrupted data in that data stripe and said parity data associated with that data stripe.

12. The computer system of claim 11, wherein the memory controller is operable to restore the corrupted data stored by said memory module by transferring the uncorrupted data and the associated parity data via exclusive-OR logic.

13. The computer system of claim 11, wherein said parity data is stored separately from said stripes of data.

14. The computer system of claim 11, wherein the memory controller includes a cache operable to store data blocks and parity data read from a data stripe in said memory array, and to utilize said cache stored data in calculating parity associated with new data blocks written to said memory array.

15. The computer system of claim 11, wherein said memory controller is operable to restore the corrupted data in a stripe of data by transferring said uncorrupted data and said parity data via exclusive-OR logic.

16. The computer system of claim 15, wherein the memory controller includes a cache operable to store data blocks and parity data read from a said stripe of data and to utilize said cache stored data in calculating parity associated with new data blocks written to said memory array.