Data caching and data-aware placement to accelerate machine learning applications
AssigneeInternational Business Machines Corporation
InventorsSeetharami R. SEELAM, Andrea REALE, Christian PINTO, Yiannis GKOUFAS, Kostas KATRINIS, Steven N. ELIUK
Life of this patent
Estimating the term…
Abstract
Embodiments for data caching and data-aware placement for machine learning by a processor. Data may be cached in a distributed data store to one or more local compute nodes of cluster of nodes with the cached data. A new job may be scheduled, according to cache and data locality awareness, on the one or more local compute nodes with the cached data needed for execution.
Claims20
A method for data caching and data-aware placement for machine learning by a processor in a multi-tenant computing environment, comprising:
- caching data in a distributed data store to one or more local compute nodes of a cluster of nodes;
- and
- scheduling, according to cache and data locality awareness, a new job on the one or more local compute nodes with the cached data needed for execution.
A system for data caching and data-aware placement for machine learning by a processor in a multi-tenant computing environment, comprising:
- one or more computers with executable instructions that when executed cause the system to:
- cache data in a distributed data store to one or more local compute nodes of a cluster of nodes;
- and
- schedule, according to cache and data locality awareness, a new job on the one or more local compute nodes with the cached data needed for execution.
A computer program product for data caching and data-aware placement for machine learning by a processor in a multi-tenant computing environment, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
- an executable portion that caches data in a distributed data store to one or more local compute nodes of a cluster of nodes;
- and
- an executable portion that schedules, according to cache and data locality awareness, a new job on the one or more local compute nodes with the cached data needed for execution.
Dependencies are inferred from the claim text.