Method and system for reducing volatile memory DRAM power budget

• US 7,483,319 B2
• Filed: 05/27/2008
• Issued: 01/27/2009
• Est. Priority Date: 02/25/2005
• Status: Active Grant

• Current Assignee: Cufer Asset Ltd LLC (Intellectual Ventures LLC)
• Original Assignee: Frankfurt GmbH LLC (Intellectual Ventures LLC)
• Inventors: Brown, Terry C.
• Primary Examiner(s): Phung; Anh
• Assistant Examiner(s): Nguyen; Hien N
• Application Number:
  US12/127,745
• Publication Number:
  US 20080225614A1
• Time in Patent Office: 245 Days
• Field of Search: 365/149, 365/200, 365/222
• US Class Current: 365/200
• CPC Class Codes: G06F 11/106 Correcting systematically all correctable errors, i.e. scrubbing

Abstract

A portable device (114) includes a power source (120), a volatile memory (118) requiring refreshing to avoid data loss, and a memory management module (116). The volatile memory (118) is operably coupled to the power source (120) for power. The memory management module (116) is operably coupled to the volatile memory (118). The memory management module (116) is also adapted to refresh the volatile memory (118) at a refresh rate which causes refresh-based errors and to correct the refresh-based errors. Also disclosed is a method for reduced power consumption by a volatile memory requiring refreshing to avoid data loss in which such a volatile memory is refreshed (122) at a refresh rate. All defective bits are detected (124) at the refresh rate. An error correction code is selected (126) for correcting the defective bits.

Claims

1. An apparatus, comprising:

a dynamic memory comprising a plurality of data storage blocks configured to store data, wherein the plurality of data storage blocks comprises a plurality of faultless data storage blocks and one or more faulty data storage blocks, each faulty data storage block having one or more faulty storage locations, and wherein the data to be stored in each faulty data storage block has an error correction code that is associated with an error correction method selected based at least in part on the amount of faulty storage locations in the faulty data storage block; and

a memory management module coupled to the dynamic memory and configured to;

read back data stored in an addressed one of the plurality of data storage blocks;

determine whether the addressed data storage block is a faulty data storage block, and if the addressed data storage block is determined to be a faulty data storage block, further;

determine the selected error correction method for the faulty data storage block, and

correct the read back data using the associated error correction code and in accordance with the determined error correction method.

2. The apparatus of claim 1, wherein the memory management module is further configured to determine the selected error correction method by performing a lookup to a memory fault table.

3. The apparatus of claim 1, wherein the memory management module is further configured to determine whether the addressed data storage block is a faulty data storage block by performing a lookup to a memory fault table.

4. The apparatus of claim 1, wherein the memory management module is further configured to:

receive write instructions to write incoming data to a second addressed data storage block; and

determine whether the second addressed data storage block is a faulty data storage block, and if so, determine a second error correction method based at least in part on a second amount of faulty storage locations in the faulty data storage block.

5. The apparatus of claim 4, wherein the memory management module is further configured to:

determine a second error correction code based at least in part on the second error correction method; and

write the incoming data to the second addressed data block, including writing the second error correction code into the dynamic memory.

6. A method, comprising

reading stored data from an addressed one of a plurality of data storage blocks of a dynamic memory, wherein the plurality of data storage blocks comprises a plurality of faultless data storage blocks and one or more faulty data storage blocks, each faulty data storage block having one or more faulty storage locations, and wherein the data stored in each faulty data storage block has an error correction code that is associated with an error correction method selected based at least in part on the amount of faulty storage locations in the faulty data storage block;

determining whether the addressed data storage block is a faulty data storage block; and

if the addressed data storage block is a faulty data storage block, determining the selected error correction method associated with the error correction code employed to store data in the addressed data storage block, and correcting the data read using the error correction code and in accordance with the determined error correction method.

7. The method of claim 6, wherein the determining whether the addressed data storage block is a faulty data storage block comprises performing a lookup to a memory fault table.

8. The method of claim 6, wherein the determining the selected error correction method comprises performing a lookup to a memory fault table.

9. A method, comprising:

receiving an instruction to write incoming data to a dynamic memory device having a plurality of data storage blocks, wherein the plurality of data storage blocks comprises a plurality of faultless data storage blocks and one or more faulty data storage blocks, each faulty data storage block having one or more faulty storage locations;

selecting one of the plurality of data blocks to write the incoming data;

determining whether the selected data storage block is a faulty data storage block, and if the selected data storage block is determined to be a faulty data storage block;

determining an error correction method for correcting the incoming data when the incoming data is read back from the dynamic memory after the incoming data has been stored into the selected data storage block;

calculating an error correction code for the incoming data in accordance with the error correction method; and

writing the incoming data and the error correction code to the selected data storage block.

10. The method of claim 9, further comprising:

reading back the written incoming data and the error correction code; and

correcting the read back data using the error correction code in accordance with the determined error correction method.

11. The method of claim 10, wherein correcting the read back data comprises determining whether the selected data storage block is a faulty data storage block, and if it is a faulty data storage block, using the error correction code in accordance with the determined error correction method.

12. The method of claim 9, wherein the determining whether the selected data storage block is a faulty data storage block comprises performing a lookup to a known defect table.

13. The method of claim 12, wherein the determining the error correction method comprises performing a second lookup either to an error correction field in the known defect table or to a separate error correction method table, cross-referenced with the known defect table.