Patents by Inventor Douglas Griffith
Douglas Griffith has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20160224427Abstract: A method for enhanced restart of a core dumping application is provided. The method includes stopping a plurality of threads in an address space, except for the thread performing the core dump. Computational segments are remapped to client segments. Each open file descriptor in the address space is closed. The application is terminated and the client segments are flushed to external storage.Type: ApplicationFiled: April 19, 2016Publication date: August 4, 2016Inventors: Anand T. Desai, Andrew Dunshea, Antonio Garcia, Douglas Griffith, Anil Kalavakolanu
-
Publication number: 20160154701Abstract: A method for enhanced restart of a core dumping application is provided. The method includes stopping a plurality of threads in an address space, except for the thread performing the core dump. Computational segments are remapped to client segments. Each open file descriptor in the address space is closed. The application is terminated and the client segments are flushed to external storage.Type: ApplicationFiled: December 2, 2014Publication date: June 2, 2016Inventors: Anand T. Desai, Andrew Dunshea, Antonio Garcia, Douglas Griffith, Anil Kalavakolanu
-
Publication number: 20160132380Abstract: A computer program product and a computer system for building a scalable system dump facility is provided. The method includes loading a component into system memory. The component includes a plurality of program modules. A component text range table entry is created for each component, whereby the component text range table entry includes: an address range, a component identifier, a data collection function, and a link to one or more related components. Upon invoking a system dump facility, a failing function instruction is determined, based on an address of the failing instruction. The component text range table is searched for an address of a failing function that is in the address range. Memory regions that are associated with the address range are transferred to a storage device first. Memory regions that are associated with related components are transferred next. Remaining memory regions are then transferred.Type: ApplicationFiled: November 7, 2014Publication date: May 12, 2016Inventors: Adekunle Bello, Douglas Griffith, Angela A. Jaehde, Robert S. Manning
-
Publication number: 20160132381Abstract: A method for building a scalable system dump facility is provided. The method includes loading a component into system memory. The component includes a plurality of program modules. A component text range table entry is created for each component, whereby the component text range table entry includes: an address range, a component identifier, a data collection function, and a link to one or more related components. Upon invoking a system dump facility, a failing function instruction is determined, based on an address of the failing instruction. The component text range table is searched for an address of a failing function that is in the address range. Memory regions that are associated with the address range are transferred to a storage device first. Memory regions that are associated with related components are transferred next. Remaining memory regions are then transferred.Type: ApplicationFiled: December 17, 2014Publication date: May 12, 2016Inventors: Adekunle Bello, Douglas Griffith, Angela A. Jaehde, Robert S. Manning
-
Patent number: 9176834Abstract: A method is provided in the illustrative embodiments. A failure is detected in a first computing node serving an application in a cluster. A subset of actions is selected from a set of actions, the set of actions configured to transfer the serving of the application from the first computing node to a second computing node in the cluster. A waiting period is set for the first computing node. The first computing node is allowed to continue serving the application during the waiting period. During the waiting period, concurrently with the first computing node serving the application, the subset of actions is performed at the second computing node. Responsive to receiving a signal of activity from the first computing node during the waiting period, the concurrent operation of the second computing node is aborted.Type: GrantFiled: September 19, 2013Date of Patent: November 3, 2015Assignee: GlobalFoundries U.S. 2 LLCInventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Patent number: 9176833Abstract: A system, and computer program product for tolerating failures using concurrency in a cluster are provided in the illustrative embodiments. A failure is detected in a first computing node serving an application in a cluster. A subset of actions is selected from a set of actions, the set of actions configured to transfer the serving of the application from the first computing node to a second computing node in the cluster. A waiting period is set for the first computing node. The first computing node is allowed to continue serving the application during the waiting period. During the waiting period, concurrently with the first computing node serving the application, the subset of actions is performed at the second computing node. Responsive to receiving a signal of activity from the first computing node during the waiting period, the concurrent operation of the second computing node is aborted.Type: GrantFiled: July 11, 2013Date of Patent: November 3, 2015Assignee: GlobalFoundries U.S. 2 LLCInventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Patent number: 9098453Abstract: A system, and computer program product for recovery in a database are provided in the illustrative embodiments. A failure is detected in a first computing node, the first computing node serving the database in a cluster of computing nodes. A snapshot is created of data of the database. A subset of log entries is applied to the snapshot, the applying modifying the snapshot to result in a modified snapshot. An access of the first computing node to the data of the database is preserved. Responsive to receiving a signal of activity from the first computing node during the applying and after a grace period has elapsed, the applying is aborted such that the first computing node can continue serving the database in the cluster.Type: GrantFiled: July 11, 2013Date of Patent: August 4, 2015Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Patent number: 9098454Abstract: A method for recovery in a database is provided in the illustrative embodiments. A failure is detected in a first computing node, the first computing node serving the database in a cluster of computing nodes. A snapshot is created of data of the database. A subset of log entries is applied to the snapshot, the applying modifying the snapshot to result in a modified snapshot. An access of the first computing node to the data of the database is preserved. Responsive to receiving a signal of activity from the first computing node during the applying and after a grace period has elapsed, the applying is aborted such that the first computing node can continue serving the database in the cluster.Type: GrantFiled: September 19, 2013Date of Patent: August 4, 2015Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Patent number: 8990541Abstract: A method, system, and computer program product for improving memory utilization of sparse pages are provided in the illustrative embodiments. A set of virtual pages is identified. Each virtual page in the set of virtual pages is a sparse virtual page. The set of virtual pages includes a first sparse virtual page and a second sparse virtual page. At least a portion of data of the first sparse virtual page in the set of virtual pages is stored in a first physical page. The first physical page belongs to a set of consolidation physical pages, and the first physical page also stores at least a portion of the data of the second sparse virtual page. The first and the second sparse pages are mapped to the first physical page.Type: GrantFiled: September 12, 2012Date of Patent: March 24, 2015Assignee: International Business Machines CorporationInventors: Adekunle Bello, Douglas Griffith, Angela Astrid Jaehde, Srinivasa Muppala Rao
-
Publication number: 20150019494Abstract: A method for recovery in a database is provided in the illustrative embodiments. A failure is detected in a first computing node, the first computing node serving the database in a cluster of computing nodes. A snapshot is created of data of the database. A subset of log entries is applied to the snapshot, the applying modifying the snapshot to result in a modified snapshot. An access of the first computing node to the data of the database is preserved. Responsive to receiving a signal of activity from the first computing node during the applying and after a grace period has elapsed, the applying is aborted such that the first computing node can continue serving the database in the cluster.Type: ApplicationFiled: September 19, 2013Publication date: January 15, 2015Applicant: International Business Machines CorporationInventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Publication number: 20150019909Abstract: A system, and computer program product for recovery in a database are provided in the illustrative embodiments. A failure is detected in a first computing node, the first computing node serving the database in a cluster of computing nodes. A snapshot is created of data of the database. A subset of log entries is applied to the snapshot, the applying modifying the snapshot to result in a modified snapshot. An access of the first computing node to the data of the database is preserved. Responsive to receiving a signal of activity from the first computing node during the applying and after a grace period has elapsed, the applying is aborted such that the first computing node can continue serving the database in the cluster.Type: ApplicationFiled: July 11, 2013Publication date: January 15, 2015Inventors: Douglas GRIFFITH, Angela Astrid JAEHDE, Matthew Ryan OCHS
-
Publication number: 20150019900Abstract: A system, and computer program product for tolerating failures using concurrency in a cluster are provided in the illustrative embodiments. A failure is detected in a first computing node serving an application in a cluster. A subset of actions is selected from a set of actions, the set of actions configured to transfer the serving of the application from the first computing node to a second computing node in the cluster. A waiting period is set for the first computing node. The first computing node is allowed to continue serving the application during the waiting period. During the waiting period, concurrently with the first computing node serving the application, the subset of actions is performed at the second computing node. Responsive to receiving a signal of activity from the first computing node during the waiting period, the concurrent operation of the second computing node is aborted.Type: ApplicationFiled: July 11, 2013Publication date: January 15, 2015Inventors: Douglas Griffith, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Publication number: 20150019901Abstract: A method is provided in the illustrative embodiments. A failure is detected in a first computing node serving an application in a cluster. A subset of actions is selected from a set of actions, the set of actions configured to transfer the serving of the application from the first computing node to a second computing node in the cluster. A waiting period is set for the first computing node. The first computing node is allowed to continue serving the application during the waiting period. During the waiting period, concurrently with the first computing node serving the application, the subset of actions is performed at the second computing node. Responsive to receiving a signal of activity from the first computing node during the waiting period, the concurrent operation of the second computing node is aborted.Type: ApplicationFiled: September 19, 2013Publication date: January 15, 2015Applicant: International Business Machines CorporationInventors: DOUGLAS GRIFFITH, Angela Astrid Jaehde, Matthew Ryan Ochs
-
Publication number: 20140075148Abstract: A method, system, and computer program product for improving memory utilization of sparse pages are provided in the illustrative embodiments. A set of virtual pages is identified. Each virtual page in the set of virtual pages is a sparse virtual page. The set of virtual pages includes a first sparse virtual page and a second sparse virtual page. At least a portion of data of the first sparse virtual page in the set of virtual pages is stored in a first physical page. The first physical page belongs to a set of consolidation physical pages, and the first physical page also stores at least a portion of the data of the second sparse virtual page. The first and the second sparse pages are mapped to the first physical page.Type: ApplicationFiled: September 12, 2012Publication date: March 13, 2014Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: ADEKUNLE BELLO, DOUGLAS GRIFFITH, ANGELA ASTRID JAEHDE, SRINIVASA MUPPALA RAO
-
Patent number: 8356099Abstract: A method, programmed medium and system are disclosed which provide for end-to-end QoS for a set of processes that comprise a workload over nfs. A set of processes that comprise a workload such as the processes of a WPAR, or an entire LPAR are given a class designation and assigned priority/limits. The data are then passed to the server which allocates resources based on the sum total of all the current classes and their priorities and/or limits. This requires re-engineering the nfs client code to be workload-aware and the nfs server code to accommodate the resource allocation and prioritization needs of the nfs clients.Type: GrantFiled: April 29, 2012Date of Patent: January 15, 2013Assignee: International Business Machines CorporationInventors: Adekunle Bello, Douglas Griffith, Somasundaram Krishnasamy, Aruna Yedavilli
-
Patent number: 8321569Abstract: A method, programmed medium and system are disclosed which provide for end-to-end QoS for a set of processes that comprise a workload over nfs. A set of processes that comprise a workload such as the processes of a WPAR, or an entire LPAR are given a class designation and assigned priority/limits. The data are then passed to the server which allocates resources based on the sum total of all the current classes and their priorities and/or limits. This requires re-engineering the nfs client code to be workload-aware and the nfs server code to accommodate the resource allocation and prioritization needs of the nfs clients.Type: GrantFiled: December 17, 2009Date of Patent: November 27, 2012Assignee: International Business Machines CorporationInventors: Adekunle Bello, Douglas Griffith, Somasundaram Krishnasamy, Aruna Yedavilli
-
Publication number: 20120215922Abstract: A method, programmed medium and system are disclosed which provide for end-to-end QoS for a set of processes that comprise a workload over nfs. A set of processes that comprise a workload such as the processes of a WPAR, or an entire LPAR are given a class designation and assigned priority/limits. The data are then passed to the server which allocates resources based on the sum total of all the current classes and their priorities and/or limits. This requires re-engineering the nfs client code to be workload-aware and the nfs server code to accommodate the resource allocation and prioritization needs of the nfs clients.Type: ApplicationFiled: April 29, 2012Publication date: August 23, 2012Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Adekunle Bello, Douglas Griffith, Somasundaram Krishnasamy, Aruna Yedavilli
-
Publication number: 20110258306Abstract: The acquisition of a lock among nodes of a divided cluster is disclosed. A method is performable by each of at least one node of the cluster. A node waits for a delay corresponding to its identifier. The node asserts intent to acquire the lock by writing its identifier to X and Y variables where another node has failed to acquire the lock. The node waits for another node to acquire the lock where the other node has written to X, and proceeds where Y remains equal to its own identifier. The node waits for another node to acquire the lock where the other node has written to a Z variable, and writes its own identifier to Z and proceeds where the other node has failed. The node writes a value to Y indicating that it is acquiring the lock, and maintains acquisition by periodically writing to Z.Type: ApplicationFiled: June 25, 2011Publication date: October 20, 2011Inventors: Sudhir G. Rao, Myung M. Bae, Thomas K. Clark, Douglas Griffith, Roger L. Haskin, Shah Mohammad Rezaul Islam, Felipe Knop, Soumitra Sarkar, Frank B. Schmuck, Theodore B. Vojnovich, Yi Zhou, Robert Curran
-
Patent number: 7991753Abstract: The acquisition of a lock among nodes of a divided cluster is disclosed. A method is performable by each of at least one node of the cluster. A node waits for a delay corresponding to its identifier. The node asserts intent to acquire the lock by writing its identifier to X and Y variables where another node has failed to acquire the lock. The node waits for another node to acquire the lock where the other node has written to X, and proceeds where Y remains equal to its own identifier. The node waits for another node to acquire the lock where the other node has written to a Z variable, and writes its own identifier to Z and proceeds where the other node has failed. The node writes a value to Y indicating that it is acquiring the lock, and maintains acquisition by periodically writing to Z.Type: GrantFiled: May 21, 2004Date of Patent: August 2, 2011Assignee: International Business Machines CorporationInventors: Sudhir G. Rao, Myung M. Bae, Thomas K. Clark, Douglas Griffith, Roger L. Haskin, Shah Mohammad Rezaul Islam, Felipe Knop, Soumitra Sarkar, Frank B. Schmuck, Theodore B. Vojnovich, Yi Zhou, Robert Curran
-
Publication number: 20110153825Abstract: A method, programmed medium and system are disclosed which provide for end-to-end QoS for a set of processes that comprise a workload over nfs. A set of processes that comprise a workload such as the processes of a WPAR, or an entire LPAR are given a class designation and assigned priority/limits. The data are then passed to the server which allocates resources based on the sum total of all the current classes and their priorities and/or limits. This requires re-engineering the nfs client code to be workload-aware and the nfs server code to accommodate the resource allocation and prioritization needs of the nfs clients.Type: ApplicationFiled: December 17, 2009Publication date: June 23, 2011Applicant: International Business Machines CorporationInventors: Adekunle Bello, Douglas Griffith, Somasundaram Krishnasamy, Aruna Yedavilli