US5754829A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Parallel computer system operating method employing a manager node to distribute attributes of individual servers
Assignee: Hitachi Software Engineering Co Ltd
•
Filing: 1995-08-24
•
Granted: 1998-05-19
•
Expires: 2015-08-24
Abstract
Abstract A parallel computer system in which a plurality of host computers are connected in a network and a plurality of servers having different attributes are allocated to the plurality of host computers. One of the plurality of host computers is previously designated as a manager node. Each of the host computers judges whether or not the own host computer is the manager node when started. The host computer decided as the manager node issues through the network to the other host computers a request demanding report of information on respective resources possessed by the other host computers on a broadcast transmission basis. In response to the broadcast transmission, the other host computers transmit information on their own resources through said network to said manager node. The manager node, on the basis of the resource information transmitted from the other host computers, determines attributes of servers to be carried out by the other host computers and transmits the determined server attributes through the network to the other host computers. Each of the other host computers receives the server attributes determined by the manager node and initializes its own host computer with the received associated server attributes.
Inventors
- 👤 Masahisa Motohiro
- 👤 Toshio Shibuya
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-31 09:01
CPC Classifications
H; H04; H04L; H04L9/00; H04L9/40
Claims Analysis
Total Claims: 21
Claim 1
Independent Claim
189 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having information as to correspondence relationship between each of server types and a desired ratio or number of servers to be allocated for a server type in question with respect to all the servers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; in a host computer determined as the manager node, determining attributes indicating types of servers to be carried out by other host computers on the basis of the information as to correspondence relationship, and transmitting the determined server attributes from said manager node to the other host computers via said network; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated host computer with the received server attributes.
Claim 2
Independent Claim
189 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having information as to correspondence relationship between each of server types and a desired ratio or number of servers to be allocated for a server type in question with respect to all the servers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; in a host computer determined as the manager node, determining attributes indicating types of servers to be carried out by other host computers on the basis of the information as to correspondence relationship, and transmitting the determined server attributes from said manager node to the other host computers via said network; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated host computer with the received server attributes.
Claim 3
Independent Claim
258 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having information as to correspondence relationship between each of server types and a desired ratio or number of servers to be allocated for a server type in question with respect to all the servers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; issuing from the host computer judged as the manager node to other host computers through said network a request demanding reports of information on respective resources possessed by the other host computers on a broadcast transmission basis; in response to said broadcast transmission, transmitting information on their own resources from the other host computers through said network to said manager node, in said manager node, on the basis of the resource information transmitted from the other host computers, and the information as to correspondence relationship, determining attributes indicating types of servers to be carried out by the other host computers and transmitting the determined server attributes from said manager node through said network to the other host computers; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated own host computer with the received associated server attributes.
Claim 4
Independent Claim
258 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having information as to correspondence relationship between each of server types and a desired ratio or number of servers to be allocated for a server type in question with respect to all the servers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; issuing from the host computer judged as the manager node to other host computers through said network a request demanding reports of information on respective resources possessed by the other host computers on a broadcast transmission basis; in response to said broadcast transmission, transmitting information on their own resources from the other host computers through said network to said manager node, in said manager node, on the basis of the resource information transmitted from the other host computers, and the information as to correspondence relationship, determining attributes indicating types of servers to be carried out by the other host computers and transmitting the determined server attributes from said manager node through said network to the other host computers; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated own host computer with the received associated server attributes.
Claim 5
Independent Claim
197 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having a server configuration definition table showing corresponding relationships between types of servers and a desired ratio of the host computers allocated by the servers with respect to all the host computers or a number of the allocated host computers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; in a host computer determined as said manager nodes determining, on the basis of the contents of said table, attributes indicating types of the respective servers to be carried out by the other host computers, and transmitting the determined server attributes from said manager node to the other host computers via said network; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated host computer with the received server attributes.
Claim 6
Independent Claim
197 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having a server configuration definition table showing corresponding relationships between types of servers and a desired ratio of the host computers allocated by the servers with respect to all the host computers or a number of the allocated host computers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; in a host computer determined as said manager nodes determining, on the basis of the contents of said table, attributes indicating types of the respective servers to be carried out by the other host computers, and transmitting the determined server attributes from said manager node to the other host computers via said network; and in each of the other host computers, receiving the server attributes determined by said manager node and initializing an associated host computer with the received server attributes.
Claim 7
Dependent Claim
65 words
A method as set forth in claim 3, further comprising the steps of: transmitting said determined server attributes from said manager node through said network to the other host computers on a broadcast basis at intervals of a predetermined time; and in response to the server attributes transmitted at said predetermined time intervals, initializing a newly started host computer with the server attributes associated therewith.
Claim 8
Dependent Claim
73 words
A method as set forth in claim 3, further comprising the steps of: transmitting said determined server attributes from said manager node through said network to the other host computers on a broadcast basis at intervals of a predetermined time; and in said manager node, monitoring presence or absence of a host computer not responding to the server attributes transmitted at said predetermined time intervals to detect an abnormality in the host computer.
Claim 9
Dependent Claim
96 words
A method as set forth in claim 5, further comprising the steps of: in said manager node, when an abnormality in the host computer is detected, detecting presence or absence of an alternate one of the host computers to the abnormal host computer; when said alternate host computer is present, transmitting server attributes of the abnormal host computer from said manager node through said network to said alternate host computer; and in said alternate host computer, receiving said server attributes determined by said manager node and initializing its own host computer with the received server attributes.
Claim 10
Dependent Claim
50 words
A method as set forth in claim 3, wherein one of the host computers has a resource in which a file indicative of its own host computer as the manager node is stored, and the host computer recognizes itself, when started, to be the manager node based on said file.
Claim 11
Dependent Claim
26 words
A method as set forth in claim 3, wherein one of the host computers recognizes itself to be the manager node when said table is initialized.
Claim 12
Independent Claim
177 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having a server configuration definition table showing corresponding relationships between types of servers and a desired ratio of the host computers allocated by the servers with respect to all the host computers or a number of the allocated host computers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; issuing from the host computer judged as the manager node to other host computers through said network a request demanding reports of information on respective resources possessed by the other host computers on a broadcast transmission basis; in response to said broadcast transmission, transmitting information on their own resources from the other host computers through said network to said manager node.
Claim 13
Independent Claim
177 words
A method for allocating a plurality of servers having different attributes to a plurality of host computers connected through a network in a parallel computer system, one of said plurality of host computers being previously designated as a manager node, the host computer previously designated as said manager node having a server configuration definition table showing corresponding relationships between types of servers and a desired ratio of the host computers allocated by the servers with respect to all the host computers or a number of the allocated host computers, comprising the steps of: judging in each of said host computers whether or not an associated host computer itself is used as said manager node when started; issuing from the host computer judged as the manager node to other host computers through said network a request demanding reports of information on respective resources possessed by the other host computers on a broadcast transmission basis; in response to said broadcast transmission, transmitting information on their own resources from the other host computers through said network to said manager node.
Claim 14
Dependent Claim
65 words
A method as set forth in claim 9, further comprising the steps of: transmitting said determined server attributes from said manager node through said network to the other host computers on a broadcast basis at intervals of a predetermined time; and in response to the server attributes transmitted at said predetermined time intervals, initializing a newly started host computer with the server attributes associated therewith.
Claim 15
Dependent Claim
73 words
A method as set forth in claim 9, further comprising the steps of: transmitting said determined server attributes from said manager node through said network to the other host computers on a broadcast basis at intervals of a predetermined time; and in said manager node, monitoring presence or absence of a host computer not responding to the server attributes transmitted at said predetermined time intervals to detect an abnormality in the host computer.
Claim 16
Dependent Claim
96 words
A method as set forth in claim 11, further comprising the steps of: in said manager node, when an abnormality in the host computer is detected, detecting presence or absence of an alternate one of the host computers to the abnormal host computer; when said alternate host computer is present, transmitting server attributes of the abnormal host computer from said manager node through said network to said alternate host computer; and in said alternate host computer, receiving said server attributes determined by said manager node and initializing its own host computer with the received server attributes.
Claim 17
Dependent Claim
56 words
A method as set forth in claim 9, wherein the information on the resources transmitted from the other host computers include host computer identifiers for identifying the other host computers, and said manager node transmits said determined server attributes to the other host computers with use of the host computer identifiers contained in said resource information.
Claim 18
Dependent Claim
50 words
A method as set forth in claim 9, wherein one of the host computers has a resource in which a file indicative of its own host computer as the manager node is stored, and the host computer recognizes itself, when started, to be the manager node based on said file.
Claim 19
Dependent Claim
26 words
A method as set forth in claim 9, wherein one of the host computers recognizes itself to be the manager node when said table is initialized.
Claim 20
Dependent Claim
175 words
A parallel computer system comprising: at least one network; a plurality of host computers connected to said at least one network; and resources possessed by said plurality of host computers, wherein one of said plurality of host computers is previously designated as a manager node; a host computer recognizing to be said manager node when started issues to other host computers through said network a request demanding report of information on respective resources possessed by the other host computers on a broadcast transmission basis; the other host computers, in response to said broadcast transmission, transmit information on their own resources through said network to said manager node; said manager node, on the basis of the resource information transmitted from the other host computers; determines attributes of servers to be carried out by the other host computers and transmits the determined server attributes through said network to the other host computers, and the other host computers receive the server attributes determined by said manager node and initialize the associated host computers with the received server attributes.
Claim 21
Dependent Claim
175 words
A parallel computer system comprising: at least one network; a plurality of host computers connected to said at least one network; and resources possessed by said plurality of host computers, wherein one of said plurality of host computers is previously designated as a manager node; a host computer recognizing to be said manager node when started issues to other host computers through said network a request demanding report of information on respective resources possessed by the other host computers on a broadcast transmission basis; the other host computers, in response to said broadcast transmission, transmit information on their own resources through said network to said manager node; said manager node, on the basis of the resource information transmitted from the other host computers; determines attributes of servers to be carried out by the other host computers and transmits the determined server attributes through said network to the other host computers, and the other host computers receive the server attributes determined by said manager node and initialize the associated host computers with the received server attributes.
References Cited (Backward Citations)
US5546540A
Cited
US5375070A
Cited
US5511208A
Cited
US5448724A
Cited
US5568605A
Cited
US5561769A
Cited
US5526358A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2001-05-18 | 2001-11-17 | 2002-05-19 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2005-05-18 | 2005-11-17 | 2006-05-19 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2009-05-18 | 2009-11-17 | 2010-05-19 | $7700.00 | Window Closed / Lapsed |