-
公开(公告)号:GB2501659A
公开(公告)日:2013-10-30
申请号:GB201315014
申请日:2012-01-13
Applicant: IBM
Inventor: GUPTA KARAN , PUCHA HIMABINDU , SARKAR PRASENJIT , SUBHRAVETI DINESH , JAIN RESHU
Abstract: Embodiments of the invention relate to block layout and block allocation in a file system to support transparency of application processing. At least one copy of an application is replicated in a write affinity region of a secondary server, and at least one copy of the application is replicated in a wide striping region across a cluster file system. When the application is subject to failure, application processing is transferred from the failure location to the write affinity copy. At the same time, the failed application is rebuilt using the wide striping replication of the application. Once the application is rebuilt, processing may return to the failed location employing the rebuilt application.
-
公开(公告)号:DE112012000282T5
公开(公告)日:2013-09-05
申请号:DE112012000282
申请日:2012-01-13
Applicant: IBM
Inventor: SARKAR PRASENJIT , JAIN RESHU , GUPTA KARAN , PUCHA HIMABINDU , SUBHRAVETI DINESH
IPC: G06F11/14
Abstract: Ausführungsformen der Erfindung betreffen Block-Layout und Blockzuweisung in einem Dateisystem zum Unterstützen von Transparenz bei Anwendungsverarbeitung. Es wird wenigstens eine Kopie einer Anwendung in einem Schreibaffinitäts-Bereich eines sekundären Servers kopiert und wenigstens eine Kopie der Anwendung in einem Wide-Striping-Bereich über ein Cluster-Dateisystem repliziert. Fällt die Anwendung aus, wird die Anwendungsverarbeitung von dem Ausfallort auf die Schreibaffinitäts-Kopie übertragen. Zur gleichen Zeit wird die ausgefallene Anwendung unter Verwendung der Wide-Striping-Replikation der wiederhergestellt. Sobald die Anwendung wiederhergestellt ist, kann die Verarbeitung unter Verwendung der wiederhergestellten Anwendung an den Ort des Anwendungsausfalls zurückkehren.
-
公开(公告)号:ES2335055T3
公开(公告)日:2010-03-18
申请号:ES07727594
申请日:2007-03-30
Applicant: IBM
Inventor: VERTES MARC , BERGHEAUD PHILIPPE , SUBHRAVETI DINESH
IPC: G06F11/20
Abstract: Una ejecución del método en un sistema informático para duplicar una ejecución de programa en dicho sistema informático que tiene unos primeros medios de almacenamiento, un destino que es accesible desde dicho sistema informático, dicho método que comprende los pasos de: - si ocurre un suceso no determinístico, bloquear la petición de un programa en ejecución correspondiente a ese suceso; - comprobar si el suceso no determinístico corresponde tanto a un mensaje que es una petición de entrada para el programa en ejecución a ser duplicado como si corresponde a una petición de salida de un programa en ejecución para enviar un mensaje de salida, denominado Suceso No Abortable o NAE que son sucesos que cambian irremediablemente el estado del mundo exterior, que no podemos hacer retroceder, - si no ocurre la notificación NAE, registrar el suceso en los primeros medios de almacenamiento, entregando el suceso al programa en ejecución a ser duplicado y desbloquear la petición de entrada; - si ocurre una notificación NAE, transferir el contenido de los primeros medios de almacenamiento al destino, esperando el reconocimiento y, cuando se recibe el reconocimiento, desbloquear la petición de salida e ir al primer paso.
-
公开(公告)号:DE602007002828D1
公开(公告)日:2009-11-26
申请号:DE602007002828
申请日:2007-03-30
Applicant: IBM
Inventor: VERTES MARC , BERGHEAUD PHILIPPE , SUBHRAVETI DINESH
IPC: G06F11/20
Abstract: A mechanism for replicating programs executing on a computer system having a first storage means is provided. The mechanism identifies the events corresponding to requests from one executing program, which may be different from the executing program to be replicated, which are non-deterministic and identifies the ‘Non Abortable Events’ (NAE's), which change irremediably the state of the external world that need to be reproduced in the replay of the programs. These events are immediately transferred for replay and the executing program is blocked until the transfer is acknowledged. For the other non-deterministic events, they are logged and sent to the executing program, the executing programs remaining blocked only if the log is full and/or if a timer between two NAEs expires, in this case a log transfer to the standby machine is performed to prepare replication before unblocking of the executing program.
-
公开(公告)号:AT445878T
公开(公告)日:2009-10-15
申请号:AT07727594
申请日:2007-03-30
Applicant: IBM
Inventor: VERTES MARC , BERGHEAUD PHILIPPE , SUBHRAVETI DINESH
IPC: G06F11/20
Abstract: A mechanism for replicating programs executing on a computer system having a first storage means is provided. The mechanism identifies the events corresponding to requests from one executing program, which may be different from the executing program to be replicated, which are non-deterministic and identifies the ‘Non Abortable Events’ (NAE's), which change irremediably the state of the external world that need to be reproduced in the replay of the programs. These events are immediately transferred for replay and the executing program is blocked until the transfer is acknowledged. For the other non-deterministic events, they are logged and sent to the executing program, the executing programs remaining blocked only if the log is full and/or if a timer between two NAEs expires, in this case a log transfer to the standby machine is performed to prepare replication before unblocking of the executing program.
-
公开(公告)号:DE112012000282B4
公开(公告)日:2022-04-28
申请号:DE112012000282
申请日:2012-01-13
Applicant: IBM
Inventor: GUPTA KARAN , PUCHA HIMABINDU , SARKAR PRASENJIT , SUBHRAVETI DINESH , JAIN RESHU
IPC: G06F11/14 , G06F16/182
Abstract: Verfahren zum Verwalten von Anwendungsverarbeitung in einem Computersystem, wobei das Verfahren die Schritte aufweist des:Konfigurierens eines Shared-Nothing-basierten Cluster-Dateisystems mit einem verteilten Satz an Serverknoten, aufweisend einen ersten Serverknoten und einen zweiten Serverknoten;Speicherns einer ersten Kopie einer ersten Anwendung lokal auf dem ersten Knoten in einem ersten Schreibaffinitäts-Bereich;Speicherns einer zweiten Kopie der ersten Anwendung lokal auf dem zweiten Serverknoten in einem zweiten Schreibaffinitäts-Bereich;Speicherns einer dritten Kopie der ersten Anwendung über Wide-Striping-Bereiche in dem Cluster;lokalen Bedienens der ersten Anwendung von dem ersten Serverknoten;in Reaktion auf einen Ausfall der ersten Anwendung, des Verlagerns der ersten Anwendung auf den zweiten Serverknoten, der durch die zweite Kopie unterstützt wird; undWiederherstellens der ersten Anwendung lokal auf dem ersten Serverknoten mit der dritten Kopie.
-
公开(公告)号:GB2499945A
公开(公告)日:2013-09-04
申请号:GB201310501
申请日:2011-11-16
Applicant: IBM
Inventor: SUBHRAVETI DINESH , SARKAR PRASENJIT , GUPTA KARAN , HASKIN ROGER LEE
IPC: G06F17/30
Abstract: Embodiments of the invention relate to data replication and block allocation in a file system to support write transactions. Regions in a cluster file system are defined to support a block allocation. Blocks in the defined regions are allocated to support the data replication. A pipeline manager is provided to schedule population of the blocks in the allocated region(s) based upon network characteristics.
-
8.
公开(公告)号:CA2672481A1
公开(公告)日:2008-06-26
申请号:CA2672481
申请日:2007-06-15
Applicant: IBM
Inventor: VERTES MARC , BERGHEAUD PHILIPPE , SUBHRAVETI DINESH
Abstract: A method, computer program and system for controlling accesses to memory by threads created by a process executing on a multiprocessor computer. A pa ge table structure is allocated for each new thread and copied from the exis ting threads. The page access is controlled by a present bit and a writable bit. Upon a page fault the access is provided to one thread. The kernel hand les the new page entry creation process and set the page present bits to zer o which creates page faults. In a second embodiment, two page table structur es are created, one for one thread having access to the address space and th e other page table structure shared by all the other threads having not acce ss to the address space.
-
-
-
-
-
-
-