System wear leveling
AssigneeHewlett Packard Enterprise Development LP
InventorsGary Smerdon, Isaac R. Nassi, David P. Reed
Life of this patent
Estimating the term…
Abstract
A technique includes receiving information relating to wear of computer nodes. Based on the information relating to the wear, the technique includes ranking the computer nodes according to respective expected remaining lifetimes of the computer nodes. The technique includes, responsive to an event that corresponds to at least one of a spare computer node being added to the system or a given computer node being removed from the system, reconfiguring the system based on the ranking.
Claims20
A non-transitory machine-readable storage medium comprising instructions that upon execution cause a system of a plurality of computer nodes to:
- execute a plurality of hyper-kernels on respective computer nodes of the plurality of computer nodes;
- responsive to the execution of the plurality of hyper-kernels, provide a guest operating system distributed across the plurality of computer nodes;
- receive information relating to wear of the computer nodes;
- based on the information relating to the wear, rank the computer nodes according to respective expected remaining lifetimes of the computer nodes;
- and
- responsive to an event corresponding to at least one of a spare computer node being added to the system or a given computer node of the plurality of computer nodes being removed from the system, reconfigure the system based on the ranking.
A distributed system comprising:
- a plurality of computer nodes;
- wherein the plurality of computer nodes comprises instructions executable to:
- execute a plurality of hyper-kernels on respective computer nodes of the plurality of computer nodes;
- responsive to the execution of the plurality of hyper-kernels, provide a guest operating system distributed across the plurality of computer nodes;
- receive an indication of an event corresponding to changing a number of the plurality of computer nodes;
- and
- responsive to the event, identify a particular computer node associated with effecting the change from a pool of candidate computer nodes based on respective wear levels associated with the candidate computer nodes.
A method of a distributed system comprising a plurality of computer nodes, comprising:
- executing, by the plurality of computer nodes, a plurality of hyper-kernels;
- providing, by the plurality of hyper-kernels, a guest operating system distributed across the plurality of computer nodes;
- receiving, by a node wear-leveling engine, information relating to wear of the computer nodes;
- based on the information relating to the wear, ranking, by the node wear-leveling engine, the computer nodes according to respective expected remaining lifetimes of the computer nodes;
- and
- responsive to an event corresponding to changing a number of the computer nodes, identifying, by the node wear-leveling engine, a particular computer node associated with effecting the change from a pool of candidate computer nodes based on the ranking.
Dependencies are inferred from the claim text.