Fault tolerance of data processing steps operating in either a parallel operation mode or a non-synchronous redundant operation mode
First Claim
1. A method of processing data in a data processor comprising a first data processing unit and a second data processing unit, the method comprising:
- performing first data processing steps in said first data processing unit concurrently with second data processing steps in said second data processing unit during a parallel operation;
andreplicating performances of third data processing steps in said first and second data processing units during a non-synchronised redundant operation, said non-synchronised redundant operation comprisingperforming an initial performance of said third data processing steps at said first data processing unit,recording first volatile data used by said first processing unit during said initial performance as recorded volatile data,performing a replicate performance of said third data processing steps starting later than said initial performance at said second data processing unit,using said recorded volatile data during said replicate performance,registering initial result data representative of results from said initial performance,comparing replicate result data representative of results from said replicate performance with said initial result data, andproducing an error signal in case of discrepancy between said replicate result data and said initial result data.
30 Assignments
0 Petitions
Accused Products
Abstract
A method of processing data in a data processor comprising at least two data processing units. The method comprises performing different data processing steps in the data processing units concurrently during a parallel operation, and replicating performances of selected identical data processing steps in the data processing units during a non-synchronised redundant operation. The non-synchronised redundant operation comprises an initial performance of the selected identical data processing steps in one of the data processing units and a replicate performance of the data processing steps starting later than the initial performance, preferably in another of the data processing units. Initial result data representative of results from the initial performance are registered, and compared with replicate result data representative of results from the replicate performance, and an error signal is produced in case of discrepancy.
31 Citations
20 Claims
-
1. A method of processing data in a data processor comprising a first data processing unit and a second data processing unit, the method comprising:
-
performing first data processing steps in said first data processing unit concurrently with second data processing steps in said second data processing unit during a parallel operation; and replicating performances of third data processing steps in said first and second data processing units during a non-synchronised redundant operation, said non-synchronised redundant operation comprising performing an initial performance of said third data processing steps at said first data processing unit, recording first volatile data used by said first processing unit during said initial performance as recorded volatile data, performing a replicate performance of said third data processing steps starting later than said initial performance at said second data processing unit, using said recorded volatile data during said replicate performance, registering initial result data representative of results from said initial performance, comparing replicate result data representative of results from said replicate performance with said initial result data, and producing an error signal in case of discrepancy between said replicate result data and said initial result data. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A data processor comprising
a first data processing unit and a second data processing unit, said first and second data processing units being arranged to perform different non-synchronised data processing steps during a parallel operation, and to replicate performance of identical data processing steps during a non-synchronised redundant operation, said data processor further comprising a comparison module operational during said non-synchronised redundant operation to record first volatile data, used by said first processing unit during an initial performance of said identical data processing steps at said first data processing unit, as recorded volatile data, register initial result data representative of results from said initial performance, use said recorded volatile data during a replicate performance of said identical processing steps, wherein said replicate performance starts later than said initial performance, produce replicate result data representative of results from said replicate performance, compare said replicate result data with said initial result data, and produce an error signal in case of discrepancy between said replicate result data and said initial result data.
-
20. A method of processing data in a data processor comprising a first data processing unit and a second data processing unit, the method comprising:
-
performing first data processing steps in said first data processing unit concurrently with second data processing steps in said second data processing unit during a parallel operation; and replicating performances of third data processing steps in said first and second data processing units during a non-synchronised redundant operation, said non-synchronised redundant operation comprising performing an initial performance of said third data processing steps in said first data processing unit, performing a replicate performance of said third data processing steps in said second data processing unit, starting later than said initial performance wherein said initial performance is not gated by said replicate performance and said replicate performance is not gated by said initial performance, registering initial result data representative of results from said initial performance, comparing replicate result data representative of results from said replicate performance with said initial result data, and producing an error signal in case of discrepancy between said replicate result data and said initial result data.
-
Specification