Claims Analysis
Total Claims: 20
Claim 1
Independent Claim
291 words
A distributed lock manager (DLM) for a clustered computer system, said clustered computer system including a plurality of processing nodes and having a plurality of application processes operating within said plurality of processing nodes, said distributed lock manager comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; and within each one of said processing nodes: a shared memory segment for storing lock status information for said one of said processing nodes, said shared memory segment being shared by said plurality of application processes to access the lock status information stored therein; a shared library providing said plurality of application processes access to said shared memory segment to request and be granted a lock if said lock is available; a DLM control program providing exception handling for any exception which occurs with respect to a lock managed within said one of said processing nodes; and a DLM driver including a communication service providing for the generation of data packets comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of data packets generated and transmitted by other ones of said processing nodes, said communication service further including means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a data packet from said receiving processing node for a predetermined period of time.
Claim 2
Dependent Claim
291 words
A distributed lock manager (DLM) for a clustered computer system, said clustered computer system including a plurality of processing nodes and having a plurality of application processes operating within said plurality of processing nodes, said distributed lock manager comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; and within each one of said processing nodes: a shared memory segment for storing lock status information for said one of said processing nodes, said shared memory segment being shared by said plurality of application processes to access the lock status information stored therein; a shared library providing said plurality of application processes access to said shared memory segment to request and be granted a lock if said lock is available; a DLM control program providing exception handling for any exception which occurs with respect to a lock managed within said one of said processing nodes; and a DLM driver including a communication service providing for the generation of data packets comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of data packets generated and transmitted by other ones of said processing nodes, said communication service further including means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a data packet from said receiving processing node for a predetermined period of time.
Claim 3
Dependent Claim
43 words
The distributed lock manager according to claim 3, wherein: said communication service includes means for limiting the transmission of additional data packets when a predetermined number of data packets have been transmitted without the receipt of acknowledge signals for said transmitted data packets.
Claim 4
Dependent Claim
297 words
A distributed lock manager (DLM) for a clustered computer system, said clustered computer system including a plurality of processing nodes and having a plurality of application processes operating within said plurality of processing nodes, said distributed lock manager comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; means for generating and transmitting an acknowledge signal in response to receipt of a data packet at a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node; and within each one of said processing nodes: a shared memory segment for storing lock status information for said one of said processing nodes, said shared memory segment being shared by said plurality of application processes to access the lock status information stored therein; a shared library providing said plurality of application processes access to said shared memory segment to request and be granted a lock if said lock is available; a DLM control program providing exception handling for any exception which occurs with respect to a lock managed within said one of said processing nodes; and a DLM driver including a communication service providing for the generation of data packets comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of data packets generated and transmitted by other ones of said processing nodes, said communication service further including means for generating and transmitting an acknowledge signal in response to receipt of a data packet from a sending processing node by a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node to said sending processing node.
Claim 5
Dependent Claim
297 words
A distributed lock manager (DLM) for a clustered computer system, said clustered computer system including a plurality of processing nodes and having a plurality of application processes operating within said plurality of processing nodes, said distributed lock manager comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; means for generating and transmitting an acknowledge signal in response to receipt of a data packet at a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node; and within each one of said processing nodes: a shared memory segment for storing lock status information for said one of said processing nodes, said shared memory segment being shared by said plurality of application processes to access the lock status information stored therein; a shared library providing said plurality of application processes access to said shared memory segment to request and be granted a lock if said lock is available; a DLM control program providing exception handling for any exception which occurs with respect to a lock managed within said one of said processing nodes; and a DLM driver including a communication service providing for the generation of data packets comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of data packets generated and transmitted by other ones of said processing nodes, said communication service further including means for generating and transmitting an acknowledge signal in response to receipt of a data packet from a sending processing node by a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node to said sending processing node.
Claim 6
Dependent Claim
186 words
In a clustered computer system including a network of processing nodes and a distributed lock manager for coordinating processing activities within the clustered computer system, the improvement comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a data packet from said receiving processing node for a predetermined period of time.
Claim 7
Dependent Claim
186 words
In a clustered computer system including a network of processing nodes and a distributed lock manager for coordinating processing activities within the clustered computer system, the improvement comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a data packet from said receiving processing node for a predetermined period of time.
Claim 8
Dependent Claim
145 words
In a clustered computer system including a network of processing nodes and a distributed lock manager for coordinating processing activities within the clustered computer system, the improvement comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for generating and transmitting an acknowledge signal in response to receipt of a data packet at a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node.
Claim 9
Dependent Claim
145 words
In a clustered computer system including a network of processing nodes and a distributed lock manager for coordinating processing activities within the clustered computer system, the improvement comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for generating and transmitting an acknowledge signal in response to receipt of a data packet at a receiving processing node, said acknowledge signal being attached to a data packet being transmitted by said receiving processing node.
Claim 10
Dependent Claim
35 words
The improvement according to claim 5, further comprising: means for limiting the transmission of additional datagrams when a predetermined number of datagrams have been transmitted without the receipt of acknowledge signals for said transmitted datagrams.
Claim 11
Dependent Claim
183 words
A datagram messaging service for a distributed lock manager implemented on a clustered computer system including a plurality of processing nodes, said datagram messaging service comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a datagram from said receiving processing node for a predetermined period of time.
Claim 12
Dependent Claim
183 words
A datagram messaging service for a distributed lock manager implemented on a clustered computer system including a plurality of processing nodes, said datagram messaging service comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for testing the integrity of a virtual circuit between a sending processing node and a receiving processing node, said means generating a periodic keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a datagram from said receiving processing node for a predetermined period of time.
Claim 13
Dependent Claim
141 words
A datagram messaging service for a distributed lock manager implemented on a clustered computer system including a plurality of processing nodes, said datagram messaging service comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for generating and transmitting an acknowledge signal in response to receipt of a datagram at a receiving processing node, said acknowledge signal being attached to a datagram being transmitted by said receiving processing node.
Claim 14
Dependent Claim
141 words
A datagram messaging service for a distributed lock manager implemented on a clustered computer system including a plurality of processing nodes, said datagram messaging service comprising: a plurality of virtual circuits, a virtual circuit between each pair of said processing nodes within said clustered computer system; a distributed lock manager driver included within each processing node, said driver including a communication service providing for the generation of datagrams comprising lock manager instructions for transmission to other ones of said processing nodes via said virtual circuits, said communication service also providing for the receipt of datagrams generated and transmitted by other ones of said processing nodes; and means for generating and transmitting an acknowledge signal in response to receipt of a datagram at a receiving processing node, said acknowledge signal being attached to a datagram being transmitted by said receiving processing node.
Claim 15
Dependent Claim
37 words
The datagram messaging system according to claim 8, further comprising: means for limiting the transmission of additional datagrams when a predetermined number of datagrams have been transmitted without the receipt of acknowledge signals for said transmitted datagrams.
Claim 16
Independent Claim
162 words
A method for operating a distributed lock manager in a clustered computer system, said clustered computer system including a plurality of processing nodes interconnected by a network, said method comprising the steps of: establishing a virtual circuit between each pair of said processing nodes within said clustered computer system; directly sending datagrams comprising lock manager instructions from a sending processing node to a receiving processing node via said virtual circuits, and subsequently directly receiving an acknowledgment signal back from said receiving processing node to said sending processing node; and testing the integrity of a virtual circuit between said sending processing node and said receiving processing node by generating a keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a datagram from said receiving processing node for a predetermined period of time.
Claim 17
Independent Claim
162 words
A method for operating a distributed lock manager in a clustered computer system, said clustered computer system including a plurality of processing nodes interconnected by a network, said method comprising the steps of: establishing a virtual circuit between each pair of said processing nodes within said clustered computer system; directly sending datagrams comprising lock manager instructions from a sending processing node to a receiving processing node via said virtual circuits, and subsequently directly receiving an acknowledgment signal back from said receiving processing node to said sending processing node; and testing the integrity of a virtual circuit between said sending processing node and said receiving processing node by generating a keepalive message packet which is periodically sent by said sending processing node to said receiving processing node requesting the transmission of an acknowledge signal in response thereto, said keepalive message packet being transmitted in the absence of the receipt of a datagram from said receiving processing node for a predetermined period of time.
Claim 18
Dependent Claim
36 words
The method according to claim 10, further comprising the step of limiting the transmission of additional datagrams when a predetermined number of datagrams have been transmitted without the receipt of acknowledge signals for said transmitted datagrams.
Claim 19
Independent Claim
132 words
A method for operating a distributed lock manager in a clustered computer system, said clustered computer system including a plurality of processing nodes interconnected by a network, said method comprising the steps of: establishing a virtual circuit between each pair of said processing nodes within said clustered computer system; directly sending datagrams comprising lock manager instructions from a sending processing node to a receiving processing node via said virtual circuits, and subsequently directly receiving an acknowledgment signal back from said receiving processing node to said sending processing node; and generating and transmitting an acknowledgment signal in response to receipt of a datagram from said sending processing node by said receiving processing node, said acknowledge signal being attached to another datagram being transmitted by said receiving processing node to said sending processing node.
Claim 20
Independent Claim
132 words
A method for operating a distributed lock manager in a clustered computer system, said clustered computer system including a plurality of processing nodes interconnected by a network, said method comprising the steps of: establishing a virtual circuit between each pair of said processing nodes within said clustered computer system; directly sending datagrams comprising lock manager instructions from a sending processing node to a receiving processing node via said virtual circuits, and subsequently directly receiving an acknowledgment signal back from said receiving processing node to said sending processing node; and generating and transmitting an acknowledgment signal in response to receipt of a datagram from said sending processing node by said receiving processing node, said acknowledge signal being attached to another datagram being transmitted by said receiving processing node to said sending processing node.