Error detection and recovery of data in striped channels

• US 7,246,303 B2
• Filed: 03/24/2003
• Issued: 07/17/2007
• Est. Priority Date: 03/25/2002
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Bansal, Akash, Varma, Anujan, Bannur, Jaisimha
• Primary Examiner(s): CHUNG, PHUNG M
• Application Number:
  US10/396,177
• Publication Number:
  US 20040017778A1
• Time in Patent Office: 1,576 Days
• Field of Search: 714/800, 714/758, 714/752, 375/260
• US Class Current: 714/800
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

In general, in one aspect, the disclosure describes an apparatus that includes a transmission module to split a data segment into a plurality of data stripes and transmit each data stripe over an associated data channel. The plurality of data channels are organized into at least one group and each group has an associated parity channel to transmit a parity stripe generated based on the data stripes within the group. The apparatus also includes a reception module to receive the plurality of data stripes and the at least one parity stripe. The apparatus further includes a controller to control the operation of the apparatus.

Claims

1. An apparatus, comprising:

a transmission module to split a data segment into a plurality of data stripes and transmit each data stripe over an associated data channel, wherein the plurality of data channels are organized into at least one group and each group has an associated parity channel to transmit a parity stripe, the parity stripe being generated based on the data stripes within the group;

a reception module to receive the plurality of data stripes over the appropriate data channels and the at least one parity stripe over the parity channel and to recreate the data segment based on the received data stripes, wherein said reception module checks for errors within each stripe, and wherein said reception module tracks number of errors per serial channel and compares to a threshold; and

a controller to control the operation of the apparatus, wherein said controller deactivates a serial channel if the number of errors in the serial channel exceeds the threshold and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes and parity stripes to either split the data segment into less data stripes or reduce number of parity stripes generated, and wherein said transmission module transmits the reconfigured combination of data stripes and parity stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured combination of data stripes and parity stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

2. The apparatus of claim 1, wherein said transmission module generates the parity stripe for a group by XORing each of the associated data stripes within the group.

3. The apparatus of claim 1, wherein said transmission module generates an error code for each stripe and transmits the error code along with the stripe.

4. The apparatus of claim 3, wherein the error code is a cyclic redundancy code (CRC).

5. The apparatus of claim 1, wherein said reception module checks for errors by computing an error code for each stripe and comparing it to a transmitted error code.

6. The apparatus of claim 1, wherein said reception module discards the data segment if more than one stripe in a group contains errors.

7. The apparatus of claim 1, wherein said reception module recreates a data stripe containing errors by XORing each of the other data stripes and the parity stripe within the group.

8. A store and forward device for striping data over a plurality of serial channels, the device comprising:

a plurality of interface cards to receive and transmit data;

a backplane connecting the plurality of cards together via a plurality of serial channels;

a transmission module to split a data segment into a plurality of data stripes and transmit each data stripe over an associated data channel, wherein the plurality of data channels are organized into at least one group and each group has an associated parity channel to transmit a parity stripe, the parity stripe being generated based on the data stripes within the group;

a reception module to receive the plurality of data stripes over the appropriate data channels and the at least one parity stripe over the parity channels, to check for errors within each stripe, and to recreate the data segment based on the received data stripes; and

a controller to control the operation of the apparatus, wherein said controller deactivates a serial channel if the number of errors in the serial channel exceeds a threshold and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes and parity stripes to either split the data segment into less data stripes or reduce number of parity stripes generated, and wherein said transmission module transmits the reconfigured combination of data stripes and parity stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured combination of data stripes and parity stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

9. The device of claim 8, wherein

said transmission module generates an error code for each stripe and transmits the error code along with the stripe;

and

said reception module checks for errors by computing an error code for each stripe and comparing it to a transmitted error code.

10. The device of claim 8, wherein said reception module discards data segment if more than one data stripe in a group contains errors.

11. The device of claim 8, wherein said interface cards are Ethernet cards.

12. An apparatus, comprising:

a transmission module to split a data segment into a plurality of data stripes, generate an error code for each data stripe, and transmit the data stripes and the associated error codes over associated serial channels;

a reception module to receive the data stripes and the associated error codes over the associated serial channels, compute error codes for each data stripe, compare the computed error codes to the transmitted error codes to detect errors, to recreate the data segment based on the received data stripes, and to track number of errors per serial channel; and

a controller to control the operation of the apparatus, wherein said controller deactivates a serial channel if the number of errors in the serial channel exceeds a threshold and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes to split the data segment into less data stripes, and wherein said transmission module transmits the reconfigured data stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured data stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

13. The apparatus of claim 12, wherein the error code computed for each stripe is a cyclic redundancy code (CRC).

14. The apparatus of claim 12, wherein said reception module recreates the data segment by combining each of the received data stripes.

15. The apparatus of claim 12, wherein the error code computed for each data segment is a cyclic redundancy code (CRC).

16. The apparatus of claim 12, wherein

said transmission module generates an error code for the data segment and transmits the data segment error code at end of the data segment;

and

said reception module generates an error code for the recreated data segment and compares it to the transmitted data segment to check for errors within the data segment.

17. The apparatus of claim 12, wherein

said transmission module organizes the plurality of data channels into at least one group and each group has an associated parity channel to transmit a parity stripe to be utilized for error correction, the parity stripe being generated based on the data stripes within the group;

and

said reception module recreates an errored data stripe by using the parity stripe and remaining data stripes within a group, if only a single data stripe within the group is errored.

18. An apparatus, comprising:

a transmission module to stripe data segments over a plurality of serial channels, wherein striping includes splitting the data segments into a plurality of data stripes, adding error detection to the data stripes, creating at least one parity stripe for the data stripes, and transmitting the plurality of data stripes and the at least one parity stripe over the plurality of serial channels;

a reception module to receive the data stripes and the at least one parity stripe over the plurality of serial channels and to recreate the data segment based on the received data stripes, wherein said reception module checks for errors within each data stripe and can correct for single data stripe errors using the parity stripe, and wherein said reception module tracks number of errors per serial channel; and

a controller to control the operation of the apparatus, wherein said controller deactivates a serial channel having the number of errors exceed a threshold and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes and parity stripes to either split the data segment into less data stripes or reduce number of parity stripes generated, and wherein said transmission module transmits the reconfigured combination of data stripes and parity stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured combination of data stripes and parity stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

19. The apparatus of claim 18, wherein said transmission module creates the parity stripe by XORing associated data stripes.

20. The apparatus of claim 18, wherein the error detection is a cyclic redundancy code (CRC).

21. The apparatus of claim 18, wherein said reception module checks for errors by computing an error code for each stripe and comparing it to a transmitted error code.

22. The apparatus of claim 18, wherein said reception module discards the data segment if more than one stripe in a group contains errors.

23. The apparatus of claim 18, wherein said reception module recreates a data stripe containing errors by XORing each of the other data stripes and the parity stripe within the group.

24. A method comprising:

striping data segments over a plurality of serial channels, wherein said striping includes splitting the data segments into a plurality of data stripes and adding error detection to the data stripes;

creating at least one parity stripe for the data stripes making up a data segment;

transmitting the parity stripe along with the data stripes;

receiving the data stripes and the at least one parity stripe over the plurality of serial channels;

checking for errors within each data stripe;

correcting single data stripe errors using the parity stripe;

recreating the data segment based on the received data stripes;

tracking number of errors per serial channel and if the number of errors exceed a threshold for a particular serial channel;

deactivating the particular serial channel;

modifying number of data stripes data segments are split into or number of parity stripes created for data segments;

transmitting the modified combination of data stripes and parity stripes over the remaining serial channels;

receiving the modified combination of data stripes and parity stripes over the remaining serial channels; and

recreating the data segment based on the received data stripes.

25. The method of claim 24, wherein said creating includes XORing associated data stripes.

26. The method of claim 24, wherein the error detection is a cyclic redundancy code (CRC).

27. The method of claim 24, wherein said checking includes computing an error code for each stripe and comparing it to a transmitted error code.

28. The method of claim 24, further comprising discarding the data segment if more than one stripe in a group contains errors.

29. The method of claim 24, further comprising recreating a data stripe containing errors by XORing each of the other data stripes and the parity stripe within the group.

30. An apparatus, comprising:

a transmission module to split a data segment into a plurality of data stripes and transmit each data strip over an associated serial channel;

a reception module to receive the plurality of data stripes over the associated serial channels, to recreate the data segment based on the received data stripes, to check the data stripes for errors, and track number of errors per serial channel; and

a controller to monitor the number of errors per serial channel and compare to a threshold, wherein if the threshold is exceeded for a certain serial channel said controller deactivates the certain serial channel and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes to split the data segment into less stripes, and wherein said transmission module transmits the reconfigured data stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured data stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

31. The apparatus of claim 30, wherein

said transmission module generates an error code for each stripe and transmits the error code along with the data stripe;

and

said reception module checks for errors by computing an error code for each data stripe and comparing it to a transmitted error code.

32. The apparatus of claim 30, wherein

said controller organizes the data stripes into at least one group;

said transmission module generates a parity stripe for each of the at least one group by XORing each of the associated data stripes within the group; and

said reception module recreates a data stripe containing errors if the data stripe is only data stripe within an associated grouping containing errors by XORing the other data stripes and the parity stripe within the group.

33. An apparatus, comprising:

a transmission module to split a data segment into a plurality of data stripes, organize the plurality of data stripes into at least one group, generate a parity stripe for each group based on the data stripes within the group, and transmit the plurality of data stripes and the at least one parity stripe over associated serial channels;

a reception module to receive the plurality of data stripes and the at least one parity stripe from the associated serial channels, recreate the data segment based on the received data stripes, check for errors in the plurality of data stripes and the at least one parity stripe, and track number of errors per serial channel; and

a controller to monitor the number of errors per serial channel and compare to a threshold, wherein if the threshold is exceeded for a certain serial channel said controller deactivates the certain serial channel and informs said transmission module and said reception module of the deactivation,

wherein said transmission module reconfigures generation of data stripes and parity stripes to either split the data segment into less data stripes or reduce number of parity stripes generated, and wherein said transmission module transmits the reconfigured combination of data stripes and parity stripes over the remaining serial channels, and

wherein said reception module receives the reconfigured combination of data stripes and parity stripes over the remaining serial channels and recreates the data segment based on the received data stripes.

34. The apparatus of claim 33, wherein

said transmission module generates an error code for each stripe and transmits the error code along with the stripe;

and

said reception module checks for errors by computing an error code for each stripe and comparing it to a transmitted error code.

35. The apparatus of claim 33, wherein

said transmission module generates the parity stripe for each of the at least one group by XORing each of the associated data stripes within the group;

and

said reception module recreates a data stripe containing errors if the data stripe is only data stripe within an associated grouping containing errors by XORing the other data stripes and the parity stripe within the group.

36. The apparatus of claim 33, wherein said transmission module generates an error code for the data segment and transmits the error code over the serial channels, wherein said reception module computes an error code for the data segment after the segment is recreated and compares to the received error code.

37. The apparatus of claim 36, wherein said reception module discards the data segment if there is an error in the data segment but not in any data stripe or if there is an error in a data stripe but not in the data segment.