Patent US6366945B1 successfully refreshed from Google Patents.
US6366945B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Flexible dynamic partitioning of resources in a cluster computing environment
Assignee: International Business Machines Corp
•
Filing: 1997-05-23
•
Granted: 2002-04-02
•
Expires: 2017-05-23
Abstract
Abstract The invention provides the mechanism, hereinafter referred to as Flexible Dynamic Partitioning (FDP), to allocate and reallocate resources among scheduling schemes of many types for multicomputing environments. Resources can include, but are not limited to, processors, disks and communications connections. Partitioning of resources can be initiated by both application and system triggers. Once dynamic partitioning is triggered, FDP allows a partition to invoke a set of resource allocation functions associated with its partition. The reallocation function performs a set of resource matchings and determine the necessary resource movement among partitions.
Inventors
- 👤 Liana Liyow Fong
- 👤 Ajei Sarat Gopal
- 👤 Nayeem Islam
- 👤 Andreas Leonidas Prodromidis
- 👤 Mark Steven Squillante
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-08 16:48
CPC Classifications
G; G06; G06F; G06F9/00; G06F9/06; G06F9/46; G06F9/50; G06F9/5061; G06F9/5066
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
127 words
A method for providing resource allocation among partitions in a partitioned multi-node computing environment comprising a plurality of partitions each applying its own scheduling methodology, comprising the steps of: dynamically detecting at least one event which may require reallocation of resources; determining if said at least one event necessitates reallocation of resources among said partitions; conducting reallocation evaluation to determine if reallocation is appropriate based on system reallocation criteria; and if reallocation is appropriate, conducting resource reallocation by instructing at least one of said plurality of partitions to perform evaluating of resource availability across its partition according to its scheduling methodology and to release at least one available resource; and allocating said at least one released resource to at least one other of said plurality of partitions.
Claim 2
Independent Claim
127 words
A method for providing resource allocation among partitions in a partitioned multi-node computing environment comprising a plurality of partitions each applying its own scheduling methodology, comprising the steps of: dynamically detecting at least one event which may require reallocation of resources; determining if said at least one event necessitates reallocation of resources among said partitions; conducting reallocation evaluation to determine if reallocation is appropriate based on system reallocation criteria; and if reallocation is appropriate, conducting resource reallocation by instructing at least one of said plurality of partitions to perform evaluating of resource availability across its partition according to its scheduling methodology and to release at least one available resource; and allocating said at least one released resource to at least one other of said plurality of partitions.
Claim 3
Dependent Claim
42 words
The method of claim 1 wherein said detecting step comprises perceiving the occurrence of a system event; and wherein said determining step comprises determining that said system event comprises a user-defined system event necessitating reallocation of resources to at least one partition.
Claim 4
Dependent Claim
42 words
The method of claim 1 wherein said detecting step comprises perceiving the occurrence of an application event; and wherein said determining step comprises determining that said application event comprises a user-defined application event necessitating reallocation of resources to at least one partition.
Claim 5
Dependent Claim
71 words
The method of claim 1 , wherein said multi-node environment comprises a hierarchical arrangement having at least one top level partition and more than one lower level partitions under said top level partition, and wherein said conducting resource allocation comprises said top level partition instructing one of said more than one lower level partitions to conduct resource allocation at lower level and wherein said allocating is performed at said lower level.
Claim 6
Dependent Claim
56 words
The method of claim 4 , wherein, in response to said instructing, said one of said more than one lower level partitions evaluates resource availability across nodes of said lower level partition; accesses at least one available node based upon said evaluating; and allocates said at least one available node to said at least one partition.
Claim 7
Dependent Claim
44 words
The method of claim 5 wherein each of said more than one lower level partitions is adapted for resource scheduling utilizing its own scheduling methodology and wherein said one of said more than one lower level partitions allocates in accordance with its scheduling methodology.
Claim 8
Dependent Claim
24 words
The method of claim 5 further comprising the step of queuing said reallocation based upon said evaluating, until at least one resource becomes available.
Claim 9
Dependent Claim
25 words
The method of claim 6 further comprising the step of said top level selecting one of said lower level partitions based upon its scheduling methodology.
Claim 10
Dependent Claim
34 words
The method of claim 1 wherein said conducting reallocation evaluation comprises determining elapsed time between a previous reallocation and said event detection and queuing said reallocation until said elapsed time matches a threshold time.
Claim 11
Independent Claim
119 words
A system for providing optimal utilization of computer resources among multiple tasks in a multi-node computing environment comprising: a plurality of computer resources; a plurality of computer nodes connected to each other and to said resources, said computer nodes being grouped into more than one partition, each scheduling according to its own scheduling methodology; at least one control entity for each of said more than one partition, said control entity being adapted to detect events requiring reallocation of resources among said more than one partitions; evaluate resource availability across nodes of said partition; access at least one available node based upon said evaluating; and allocate said at least one available node to one of said more than one partition.
Claim 12
Independent Claim
119 words
A system for providing optimal utilization of computer resources among multiple tasks in a multi-node computing environment comprising: a plurality of computer resources; a plurality of computer nodes connected to each other and to said resources, said computer nodes being grouped into more than one partition, each scheduling according to its own scheduling methodology; at least one control entity for each of said more than one partition, said control entity being adapted to detect events requiring reallocation of resources among said more than one partitions; evaluate resource availability across nodes of said partition; access at least one available node based upon said evaluating; and allocate said at least one available node to one of said more than one partition.
Claim 13
Dependent Claim
30 words
The system of claim 10 wherein said computer nodes are grouped into at least one top level partition and more than one lower level partitions under said top level partition.
Claim 14
Dependent Claim
31 words
The system of claim 11 wherein said computer nodes are further grouped into a plurality of successively smaller lower level partitions with the smallest lower level partitions comprising single node partitions.
Claim 15
Dependent Claim
31 words
The system of claim 11 wherein said control entity of each of said more than one lower level partitions is adapted to provide resource scheduling according to a different scheduling methodology.
Claim 16
Dependent Claim
32 words
The system of claim 12 wherein said control entity of each of said plurality of successively smaller lower level partitions is adapted to provide resource scheduling according to a different scheduling methodology.
Claim 17
Dependent Claim
24 words
The system of claim 10 further comprising at least one system administrator for establishing standards for events requiring reallocation, as applied in said evaluating.
Claim 18
Dependent Claim
31 words
The system of claim 10 wherein said control entity further comprises clock means for determining elapsed time between successive reallocations and means for comparing said elapsed time to a preset threshold.
Claim 19
Dependent Claim
31 words
The system of claim 16 wherein said control entity further comprises means for temporarily storing event information until said means for determining determines that said elapsed time exceeds said preset threshold.
Claim 20
Dependent Claim
24 words
The system of claim 10 wherein said control entity further comprises means for temporarily storing event information until said evaluating step locates available resources.
Claim 21
Dependent Claim
13 words
The system of claim 10 wherein said multi-node environment comprises, an object-oriented framework.
Claim 22
Dependent Claim
28 words
The method of claim 1 wherein said conducting reallocation evaluation comprises determining the number of waiting applications and queuing said reallocation until said number reaches a preset threshold.
References Cited (Backward Citations)
US4603382A
Cited
US5230065A
Cited
US5394547A
Cited
US5325526A
Cited
US5640604A
Cited
US5748892A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2005-04-01 | 2005-10-01 | 2006-04-02 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2009-04-01 | 2009-10-01 | 2010-04-02 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2013-04-01 | 2013-10-01 | 2014-04-02 | $7700.00 | Window Closed / Lapsed |