Systems and methods for providing continuous memory redundancy, availability, and serviceability using dynamic address space mirroring
An information handling system may include one or more processors, a memory system communicatively coupled to the one or more processors, and a program of instructions embodied in non-transitory computer readable media and configured to, when read and executed by the one or more processors, create operating system level-mirroring of address spaces for data associated with one or more processes executing on the one or more processors and dynamically reallocate address spaces used for mirroring of the data for a process of the one or more processes from a first address space to a second address space responsive to a determination that a number of correctable bit errors of a memory page associated with the first address space exceeds a threshold.
1. An information handling system comprising:
one or more processors;
a memory system communicatively coupled to the one or more processors; and
a program of instructions embodied in non-transitory computer readable media and configured to, when read and executed by the one or more processors:
receive a user selection of one or more processes tagged for mirroring;
in response to the user selection, create operating system-level mirroring of address spaces for data associated with the one or more processes executing on the one or more processors; and
dynamically reallocate address spaces used for mirroring of the data for a process of the one or more processes from a first address space to a second address space responsive to a determination that a number of correctable bit errors of a memory page associated with the first address space exceeds a threshold.
2. The information handling system of claim 1 , wherein creating operating system-level mirroring of address spaces for data associated with one or more processes executing on the one or more processors comprises, for each process of the one or more processes, selecting an address space within a non-uniform memory access (NUMA) memory domain of a processor of the one or more processors executing such process.
3. The information handling system of claim 1 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises selecting the second address space based on proximity of the second address space to a processor of the one or more processors executing the process.
4. The information handling system of claim 1 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises:
identifying a plurality of candidate address spaces for the second address space; and
selecting the second address space from the plurality of candidate address spaces based on a respective number of correctable bit errors that have occurred within each of the plurality of candidate address spaces.
5. A method comprising, in an information handling system comprising one or more processors and a memory system communicatively coupled to the one or more processors:
receiving a user selection of one or more processes tagged for mirroring;
in response to the user selection, creating operating system-level mirroring of address spaces for data associated with one or more processes executing on the one or more processors; and
dynamically reallocating address spaces used for mirroring of the data for a process of the one or more processes from a first address space to a second address space responsive to a determination that a number of correctable bit errors of a memory page associated with the first address space exceeds a threshold.
6. The method of claim 5 , wherein creating operating system-level mirroring of address spaces for data associated with one or more processes executing on the one or more processors comprises, for each process of the one or more processes, selecting an address space within a non-uniform memory access (NUMA) memory domain of a processor of the one or more processors executing such process.
7. The method of claim 5 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises selecting the second address space based on proximity of the second address space to a processor of the one or more processors executing the process.
8. The method of claim 5 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises:
identifying a plurality of candidate address spaces for the second address space; and
selecting the second address space from the plurality of candidate address spaces based on a respective number of correctable bit errors that have occurred within each of the plurality of candidate address spaces.
9. An article of manufacture comprising:
a non-transitory computer-readable medium; and
computer-executable instructions carried on the computer-readable medium, the instructions readable by a processor, the instructions, when read and executed, for causing the processor to, in an information handling system comprising one or more processors and a memory system communicatively coupled to the one or more processors:
receive a user selection of one or more processes tagged for mirroring;
in response to the user selection, create operating system-level mirroring of address spaces for data associated with one or more processes executing on the one or more processors; and
dynamically reallocate address spaces used for mirroring of the data for a process of the one or more processes from a first address space to a second address space responsive to a determination that a number of correctable bit errors of a memory page associated with the first address space exceeds a threshold.
10. The article of claim 9 , wherein creating operating system-level mirroring of address spaces for data associated with one or more processes executing on the one or more processors comprises, for each process of the one or more processes, selecting an address space within a non-uniform memory access (NUMA) memory domain of a processor of the one or more processors executing such process.
11. The article of claim 9 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises selecting the second address space based on proximity of the second address space to a processor of the one or more processors executing the process.
12. The article of claim 9 , wherein dynamically reallocating an address space used for mirroring of data for the process of the one or more processes comprises:
identifying a plurality of candidate address spaces for the second address space; and
selecting the second address space from the plurality of candidate address spaces based on a respective number of correctable bit errors that have occurred within each of the plurality of candidate address spaces.