摘要:
A method for decoding encoded data comprising integrated data and header protection is disclosed herein. In one embodiment, such a method includes receiving an extended data array. The extended data array includes a data array organized into rows and columns, headers appended to the rows of the data array, column ECC parity protecting the columns of the data array, and row ECC parity protecting the rows and headers combined. The method then decodes the extended data array. Among other operations, this decoding step includes checking the header associated with each row to determine whether the header is legal. If the header is legal, the method determines the contribution of the header to the corresponding row ECC parity. The method then reverses the contribution of the header to the corresponding row ECC parity. A corresponding apparatus (i.e., a tape drive configured to implement the above-described method) is also disclosed herein.
摘要:
A method for integrating data and header protection in tape drives includes receiving an array of data organized into rows and columns. The array is extended to include one or more headers for each row of data in the array. The method provides two dimensions of error correction code (ECC) protection for the data in the array and a single dimension of ECC protection for the headers in the array. A corresponding apparatus is also disclosed herein.
摘要:
A method for integrating data and header protection in tape drives includes receiving an array of data organized into rows and columns. The array is extended to include one or more headers for each row of data in the array. The method provides two dimensions of error correction code (ECC) protection for the data in the array and a single dimension of ECC protection for the headers in the array. A corresponding apparatus is also disclosed herein.
摘要:
Conventional C2 coding and interleaving for multi-track data tape in LTO-3/4 do not support recording data onto a number of concurrent tracks which is not a power of two. Higher-rate longer C2 codes, which do not degrade error rate performance, are provided. An adjustable format and interleaving scheme accommodates future tape drives in which the number of concurrent tracks is not necessarily a power of two. A data set is segmented into a plurality of unencoded subdata sets and parity bytes are generated for each row and column. The parameters of the C2 code include N2 as the least common multiple of the number of possible tracks to which codeword objects are to be written. COs are formed from N2 C1 codewords, mapped onto a logical data track according to information within headers of the CO and modulation encoded into synchronized COs which are written to the tape.
摘要:
For writing data to multi-track tape, a received data set is received and segmented into unencoded subdata sets, each comprising an array having K2 rows and K1 columns. For each unencoded subdata set, N1-K1 C1-parity bytes are generated for each row and N2-K2 C2-parity bytes are generated for each column. The C1 and C2 parity bytes are appended to the ends of the row and column, respectively, to form encoded C1 and C2 codewords, respectively. All of the C1 codewords per data set are endowed with a specific codeword header to form a plurality of partial codeword objects (PCOs). Each PCO is mapped onto a logical data track according to information within the header. On each logical data track, adjacent PCOs are merged to form COs which are modulation encoded and mapped into synchronized COs. Then T synchronized COs are written simultaneously to the data tape where T is the number of concurrent active tracks on the data tape.
摘要:
Conventional C2 coding and interleaving for multi-track data tape in LTO-¾ do not support recording data onto a number of concurrent tracks which is not a power of two. Higher-rate longer C2 codes, which do not degrade error rate performance, are provided. An adjustable format and interleaving scheme accommodates future tape drives in which the number of concurrent tracks is not necessarily a power of two. A data set is segmented into a plurality of unencoded subdata sets and parity bytes are generated for each row and column. The parameters of the C2 code include N2 as the least common multiple of the number of possible tracks to which codeword objects are to be written. COs are formed from N2 C1 codewords, mapped onto a logical data track according to information within headers of the CO and modulation encoded into synchronized COs which are written to the tape.
摘要:
For writing data to multi-track tape, a received data set is received and segmented into unencoded subdata sets, each comprising an array having K2 rows and K1 columns. For each unencoded subdata set, N1−K1 C1-parity bytes are generated for each row and N2−K2 C2-parity bytes are generated for each column. The C1 and C2 parity bytes are appended to the ends of the row and column, respectively, to form encoded C1 and C2 codewords, respectively. All of the C1 codewords per data set are endowed with a specific codeword header to form a plurality of partial codeword objects (PCOs). Each PCO is mapped onto a logical data track according to information within the header. On each logical data track, adjacent PCOs are merged to form COs which are modulation encoded and mapped into synchronized COs. Then T synchronized COs are written simultaneously to the data tape where T is the number of concurrent active tracks on the data tape.
摘要:
A method for efficiently rewriting data to tape is disclosed herein. In one embodiment, such a method includes writing a data set to tape, the data set comprising S sub data sets of fixed size, each sub data set comprising N code word interleaves (CWIs). The method further includes reading the data set while writing it to the tape to identify faulty CWIs. While reading the data set, the method buffers the faulty CWIs (such as by storing, identifying, and/or marking the faulty CWIs) for later retrieval. When the end of the data set is reached, the method writes corrected versions of the faulty CWIs to the end of the data set. A corresponding apparatus is also disclosed and claimed herein.
摘要:
A method for physically laying out data on tape is disclosed herein. In one embodiment, such a method includes receiving a data set, wherein the data set includes S sub data sets (SDSs) of fixed size and each SDS includes N codeword interleaves (CWIs). The method further distributes the CWIs for the S SDSs across T tracks on a physical tape medium such that the distances between CWIs of the same SDS are substantially maximized on the physical tape medium. To maximize the distances, the method periodically rotates the tracks within the data set by a track rotation value R, wherein the number of tracks T is equal to 2k, and the track rotation value R is equal to 2k−1−1. A corresponding apparatus is also disclosed herein.
摘要:
A method for decoding encoded data comprising integrated data and header protection is disclosed herein. In one embodiment, such a method includes receiving an extended data array. The extended data array includes a data array organized into rows and columns, headers appended to the rows of the data array, column ECC parity protecting the columns of the data array, and row ECC parity protecting the rows and headers combined. The method then decodes the extended data array. Among other operations, this decoding step includes checking the header associated with each row to determine whether the header is legal. If the header is legal, the method determines the contribution of the header to the corresponding row ECC parity. The method then reverses the contribution of the header to the corresponding row ECC parity. A corresponding apparatus (i.e., a tape drive configured to implement the above-described method) is also disclosed herein.