Multi-core engine for detecting bit errors

• US 8,898,536 B2
• Filed: 04/27/2007
• Issued: 11/25/2014
• Est. Priority Date: 04/27/2007
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Kogelnik, Christoph
• Primary Examiner(s): Torres, Joseph D
• Application Number:
  US11/741,016
• Publication Number:
  US 20090327818A1
• Time in Patent Office: 2,769 Days
• Field of Search: 714/758, 714/763
• US Class Current: 714/758
• CPC Class Codes: G06F 11/1004 to protect a block of data words, e.g. CRC or checksum G06F11/1076 takes precedence; security arrangements for protecting computers or computer systems against unauthorized activity G06F21/00
• CPC Class Codes: G06F 21/602 Providing cryptographic facilities or services
• CPC Class Codes: G06F 2221/2107 File encryption
• CPC Class Codes: H03M 13/09 Error detection only, e.g. using cyclic redundancy check [CRC] codes or single parity bit
• CPC Class Codes: H04L 2463/062 applying encryption of the keys
• CPC Class Codes: H04L 63/0428 wherein the data content is protected, e.g. by encrypting or encapsulating the payload

Abstract

The following description includes a method and a system of detecting bit errors in a multi-core processor. When a subatomic particle, or other matter, impacts the processing cores of the processor, bit flips may occur. To detect these bit flips and thereby prevent erroneous results, operations performed by one core are inversely performed by another core. By comparing the results of the original operation and the inverse operation, embodiments of the invention can detect errors in binary data. If an error is detected, then the operations are performed again. Alternatively, multiple cores do not perform inverse operations, but instead perform identical operations in parallel. The results from the parallel operations are compared and if the results are not identical, then the operations are repeated.

Claims

1. A computer system for detecting errors during processing, comprising:

a network adapter of the computer system configured to receive a data stream including data from a client; and

a processor of the computer system configured to process the data, the processor including one or more processor cores, wherein the one or more processor cores includes at least one of an encryption core and a decryption core, the one or more processor cores further configured to;

calculate a first signature associated with the data,

perform a first operation to the data to produce first updated data,

perform a second operation to the first updated data to produce second updated data,

calculate a second signature associated with the second updated data,

store the first signature and the second signature in at least one buffer of the computer system;

compare the first signature and the second signature from the at least one buffer,

in response to the first signature and the second signature matching, store the first updated data in a memory of the computer system, and

in response to the first signature and the second signature not matching;

(i) perform the first operation to the data to produce new first updated data, (ii) perform the second operation to the new first updated data to produce new second updated data, and (iii) calculate a new second signature associated with the new second updated data, wherein (i)-(iii) are performed until the first signature associated with the data matches the new second signature.

2. The computer system of claim 1, wherein the first operation comprises compression, and the second operation comprises decompression.

3. The computer system of claim 1, wherein the one or more processor cores are the same.

4. The computer system of claim 3, wherein the first signature comprises one of parity information, a cyclic redundancy code, a hash function computation, and a linear feedback shift register computation.

5. The computer system of claim 1, wherein the first signature is calculated before the first operation is performed and the second operation is performed after the second signature is calculated.

6. The computer system of claim 1, wherein the first operation is performed after the first signature is calculated and the second operation is performed before the second signature is calculated.

7. The computer system of claim 1, wherein the first operation comprises encryption and the second operation comprises decryption.

8. The computer system of claim 1, wherein the first operation and the second operation are performed in parallel.

9. The computer system of claim 1, wherein the first operation is an inverse operation to the second operation.

10. The computer system of claim 1, wherein the data is stored on a storage device operatively connected to the computer system if the first signature and the second signature match.

11. The computer system of claim 1, wherein a match between the first signature and the second signature indicates an absence of an error.

12. The computer system of claim 1, wherein at least one of the one or more processor cores comprises a field programmable gate array.

13. The computer system of claim 1, wherein the processor further includes the at least one buffer.

14. A method, comprising:

receiving, at a network adapter, a data stream including data from a client;

processing the data by a computer processor having one or more processor cores that includes at least one of an encryption core and a decryption core, the one or more processor cores configured for;

calculating a first signature associated with the data;

performing a first operation to the data to produce first updated data;

performing a second operation to the first updated data to produce second updated data;

calculating a second signature associated with the second data;

storing the first signature and the second signature in at least one buffer;

comparing the first signature and the second signature from the at least one buffer;

storing the first updated data in a memory coupled to the computer processor in response to the first signature and the second signature matching; and

in response to the first signature and the second signature not matching;

(i) performing the first operation to the data to produce new first updated data, (ii) performing the second operation to the new first updated data to produce new second updated data, and (iii) calculating a new second signature associated with the new second updated data, wherein (i)-(iii) are performed until the first signature associated with the data matches the new second signature.

15. The method of claim 14, wherein the first operation comprises compression, and the second operation comprises decompression.

16. The method of claim 14, wherein a first processor core and a second processor core are the same.

17. The method of claim 14, wherein the first signature comprises a fingerprint.

18. The method of claim 14, further comprising:

performing the first operation before calculating the first signature; and

performing the second operation after calculating the second signature.

19. The method of claim 14, further comprising:

performing the first operation after calculating the first signature; and

performing the second operation before calculating the second signature.

20. The method of claim 14, wherein the first operation comprises encryption and the second operation comprises decryption.

21. The method of claim 14, further comprising performing the first operation and the second operation in parallel.

22. The method of claim 14, wherein the first operation is an inverse operation to the second operation.

23. The method of claim 14, further comprising:

performing the first operation by a first processor core of the one or more processor cores; and

performing the second operation by a second processor core of the one or more processor cores.

24. The method of claim 14, further comprising performing the first operation and the second operation on the data a second time in response to detecting an error.

25. The method of claim 14, further comprising detecting an error in response to not a match between the first signature and the second signature.

26. The method of claim 14, wherein at least one of the one or more processor cores comprises an application specific integrated circuit.

27. A non-transitory computer readable medium containing executable program instructions to be executed by a processor, the computer readable medium comprising:

program instructions that receive a data stream including data from a client;

program instructions that process the data by a processor including one or more processor cores that includes at least one of an encryption core and a decryption core;

program instructions that produce a first signature associated with the data by a first processor core of the one or more processor cores;

program instructions that perform a first operation to the data to produce first updated data;

program instructions that perform a second operation to the first updated data to produce second updated data;

program instructions that produce a second signature associated with the second updated data by a second processor core of the one or more processor cores;

program instructions that store the first signature and the second signature in at least one buffer;

program instructions that compare, from the at least one buffer, the first signature and the second signature to detect whether a bit flip error associated with the data occurred when performing the first operation or the second operation, wherein the bit flip error occurs if the first signature and the second signature do not match;

program instructions that transmit the data to a location to be stored in response to the detecting that the bit flip error has not occurred; and

program instructions that, in response to the detecting that the bit flip error has occurred;

(i) perform the first operation to the data to produce new first updated data, (ii) perform the second operation to the new first updated data to produce new second updated data, and (iii) produce a new second signature associated with the new second updated data, wherein (i)-(iii) are repeated until the first signature matches the new second signature.

28. A system to detect errors during data processing, comprising:

a network adapter of the computer system configured to receive a data stream including a plurality of data blocks from a client; and

a processor of the computing system configured to process the plurality of data blocks, the processor including one or more processor cores that includes at least one of an encryption core and a decryption core;

at least one of the one or more processor cores configured to perform a first operation on a first block of the plurality of blocks to produce a first signature;

at least one of the one or more processor cores configured to perform a first function on the first data block to produce a second data block;

at least one of the one or more processor cores configured to perform a second function on the second data block to produce a third data block;

at least one of the one or more processor cores configured to perform a second operation on the third data block to produce a second signature;

a buffer configured to store the first signature and the second signature;

at least one of the one or more processor cores configured to compare the first signature and the second signature from the buffer to determine an occurrence of a bit flip error associated with the first data block that occurs when performing the first operation or the second operation, wherein the bit flip error occurs if the first signature and the second signature do not match; and

a memory configured to store the first signature and the second signature in response to the first signature and the second signature matching;

at least one of the one or more processor cores, in response to the first signature and the second signature not matching, configured to;

(i) perform the first function on the first data block to produce a new second data block, (ii) perform the second function on the new second data block to produce a new third data block, and (iii) perform the second operation on the new third data block to produce a new second signature, wherein (i)-(iii) are repeated until the first signature matches the new second signature.

29. The system of claim 28, wherein the at least one of the one or more processor cores that performs the first function and the at least one of the one or more processor cores that perform the second function are the same.