Systems and Methods for Efficient Data Preprocessing of Machine Learning Workloads
US-2024403138-A1 · Dec 5, 2024 · US
US9678802B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9678802-B2 |
| Application number | US-201213571760-A |
| Country | US |
| Kind code | B2 |
| Filing date | Aug 10, 2012 |
| Priority date | Aug 9, 2012 |
| Publication date | Jun 13, 2017 |
| Grant date | Jun 13, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A distributed node service management system utilizes multiple existing processor nodes of a distributed computing system, in support of the primary data processing functions of the distributed computing system. The distributed node service management system coordinates and manages service functions on behalf of processor nodes of the distributed computing system. Other features and aspects may be realized, depending upon the particular application.
Opening claim text (preview).
What is claimed is: 1. A computer program product for managing a distributed computing system having a plurality of processor nodes, the computer program product comprising a non-transitory computer readable storage medium having computer readable program code embodied therein for execution by a processor to perform managing operations, the managing operations comprising: performing data processing functions at each of the plurality of processor nodes including performing data processing functions at a first processor node in a first data processing role, in the distributed computing system having said plurality of processor nodes; selecting a first mode of operation of a distributed node service management system by said plurality of processor nodes in said distributed computing system including said first processor node; and in response to selection of a first mode of operation of the distributed node service management system, configuring said first processor node to also perform distributed node service management in a first distributed node service management role as a function of the selected first mode of operation, wherein the distributed node service management in the first distributed node service management role at said first processor node includes processing a service request for servicing the distributed computing system including forwarding an error notification to a support center external to the distributed computing system, wherein said configuring configures said first processor node to perform distributed node service management in a master distributed node service management role which includes joining additional processor nodes in said distributed computing system to said distributed node service management system, wherein performing distributed node service management by the first processor node in the master distributed node service management role further includes qualifying a second processor node to determine if the second processor node is qualified to join the distributed node service management system before joining the second processor node to the distributed node service management system, and wherein the qualifying a second processor node to determine if the second processor node is qualified to join the distributed node service management system, is a function of a comparison of versions of at least one of hardware and software maintained at the second processor node, relative to versions of at least one of hardware and software maintained at other processor nodes previously joined to the distributed node service management system. 2. The computer program product of claim 1 wherein performing distributed node service management by the first processor node in the master distributed node service management role further includes providing a join lock to a second processor node to prevent other processor nodes from joining the distributed node service management system while the second processor node has the join lock. 3. The computer program product of claim 1 wherein the joining additional processor nodes in the distributed computing system to the distributed node service management system includes assigning a distributed node service management role to a processor node joining the distributed node service management system. 4. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is a function of preselected user preferences. 5. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is one of a master, standby, and member distributed node service management role. 6. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is a function of preselected user preferences. 7. The computer program product of claim 6 wherein the distributed computing system comprises clusters in which each cluster includes at least two of the plurality of processor nodes of the distributed computing system, and wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is one of a master, standby, and member distributed node service management role and is a function of a preselected user preference that each cluster of the distributed computing system have at least one processor node perform distributed node service management in a standby distributed node service management role. 8. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is one of a master, standby, and member distributed node service management role and is a function of a preselected user preference that the number of processor nodes operating in a standby distributed node service management role meet or exceed a preselected percentage of the total processor nodes joined to the distributed node service management system. 9. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is selected by a user. 10. The computer program product of claim 3 wherein the distributed node service management role assigned to a processor node joining the distributed node service management system is a function of capabilities of the processor node joining the distributed node service management system. 11. The computer program product of claim 1 wherein the first processor node performing distributed node service management in the master distributed node service management role further includes obtaining a network IP address for addressing the first processor node within a network providing communication paths among the processor nodes in the distributed computing system. 12. The computer program product of claim 1 further comprising in response to the selection of the first mode of operation of the distributed node service management system, configuring additional processor nodes of the distributed computing system to also perform distributed node service management in distributed node service management roles as a function of the selected first mode of operation, wherein the configuring of processor nodes is performed independently of data processing functions at the processor nodes being configured so that the data processing functions at the processor nodes being configured are not disrupted by the configuring of the processor nodes to perform distributed node service management. 13. The computer program product of claim 12 further comprising selecting a second mode of operation of a distributed node service management system by said plurality of processor nodes in said distributed computing system, and in response to the selection of the second mode of operation of the distributed node service management system, reconfiguring processor nodes of the distributed computing system to perform distributed node service management in distributed node service management roles as a function of the selected second mode of operation, wherein the reconfiguring of processor nodes is performed independently of data processing functions at the processor nodes being reconfigured so that the data processing functions at the processor nodes being reconfigured are not disrupted by the reconfiguring of the processor nodes to perform distributed node ser
in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title
Partitioning or combining of resources · CPC title
Discovery or management thereof, e.g. service location protocol [SLP] or web services · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.