Abstract: A system including a node, wherein the node includes two separate controllers, each of which is configured to output data to a bus, or receive data from a bus, or output data to and receive data from a bus. At least one controller is configured to monitor the output of the other controller and is configured such that if the at least one controller determines that the other controller is providing improper data or signals, at least part of the output data of the other controller is nullified, overridden or superseded by an output from the at least one controller.

Abstract: Power supply voltage of a PCI or a similar communication bus interface is separated from one or more other power supply voltages on a backplane, on boards insertable into the backplane, and on the bus interface components of the boards. The power supply of the bus interface (VIO) is provided to cold spare boards inserted into the backplane, while the other voltages are not provided to the cold spare boards. Availability of the VIO on the cold spare boards prevents the VIO clamping diodes on the PCI I/O lines from grounding the PCI bus. Cold spare capability is thus provided to systems with PCI and similar communication buses.

Abstract: Provided is a method, system, and program for processing Input/Output (I/O) requests to a storage network including at least one storage device and at least two adaptors, wherein each adaptor is capable of communicating I/O requests to the at least one storage device. An error is detected in a system including a first adaptor, wherein the first adaptor is capable of communicating on the network after the error is detected. In response to detecting the error, a master switch timer is started that is less than a system timeout period if the first adaptor is the master. An error recovery procedure in the system including the first adaptor would be initiated after the system timeout period has expired. An operation is initiated to designate another adaptor in the storage network as the master if the first adaptor is the master in response to detecting an expiration of the master switch timer.

Abstract: A method for operating a system comprising multiple resources. The method comprises identifying for each resource a set of one or more failure risks for that resource. For each identified failure risk, a likelihood of failure is combined with an expected resolution time to provide a risk weight for the identified failure risk. For each resource, the risk weights for each failure risk are accumulated to provide an accumulated risk weight for the resource. A resource manager provisions workload across the multiple resources based on the accumulated risk weights for each resource.

Abstract: A method for accessing shared memory cards from each of plural processor cards is disclosed. The shared memory cards are composed of a shared memory card of an operating system and a shared memory card of a standby system in a redundant configuration, and each of plural processor cards individually access the shared memory cards. Each of the plural processor cards is connected to the shared memory card of the operating system and the shared memory card of the standby system in a point-to-point structure via corresponding serial buses and executes data transmission on a one to one basis to/from the shared memory card of the operating system and the shared memory card of the standby system.

Abstract: A process control system is provided having a plurality of I/O devices in communication using a bus. A primary redundant I/O device and a secondary redundant I/O device are coupled to the bus, where the secondary redundant I/O device is programmed to detect a primary redundant I/O device fault. The secondary redundant I/O device, upon detecting the primary redundant I/O device fault, publishes a primary redundant I/O device fault message on the bus. The controller may deactivate the primary redundant I/O device and activate the secondary redundant I/O device responsive to the primary redundant I/O device fault message.

Abstract: Provided is a system for processing Input/Output (I/O) requests to a storage network including at least one storage device and at least two adaptors, wherein each adaptor is capable of communicating I/O requests to the at least one storage device. An error is detected in a system including a first adaptor, wherein the first adaptor is capable of communicating on the network after the error is detected. In response to detecting the error, a master switch timer is started that is less than a system timeout period if the first adaptor is the master. An error recovery procedure in the system including the first adaptor would be initiated after the system timeout period has expired. An operation is initiated to designate another adaptor in the storage network as the master if the first adaptor is the master in response to detecting an expiration of the master switch timer.

Abstract: A computer system and method for allowing either a primary computer or a backup computer to access a hardware resource, depending on which computer is executing an application that needs the hardware resource. A switch is connected to the hardware resource, the primary computer and the backup computer such that in one state, the switch connects the hardware resource to the primary computer, and in another state, the switch connects the hardware resource to the backup computer. Initially, the primary computer executes the application using the hardware resource. However, when the primary computer ceases to execute the application, the switch connects the hardware resource to the backup computer, and the backup computer executes the application or copy thereof using the hardware resource.

Abstract: Embodiments of the present invention provide a system for executing program code on a processor. In these embodiments, the processor is configured to start by using a primary strand to execute program code. Upon detecting a predetermined condition, the processor is configured to instantaneously checkpoint an architectural state of the primary strand and then use the subordinate strand to copy the checkpointed state to memory while using the primary strand to continue executing the program code without interruption.

Abstract: A switch control system including a storage unit, a switch which logically sets a network topology between the storage unit and a plurality of computers, and a management server which communicates with the switch and the storage unit, wherein the storage unit includes at least one disk; wherein the management server comprises a memory and a processor, wherein the memory holds the network topology which is set by the switch, wherein when a failure is detected in one of the computers currently being used, the processor of the management server refers to the memory to change the network topology for the computer where the failure is detected and another computer which substitutes the computer where the failure is detected, and instructs the switch with the changed network topology so as to cause the switch to logically set the changed network topology, and wherein the management server controls the disk of the computer where the failure is detected to be accessible.

Abstract: A system of circumvention and recovery in a multi-function system. The system includes a plurality of peripherals, at least one event detector and a controller. The plurality of peripherals have different levels of susceptibility to environmental events. The at least one event detector is configured to detect environmental events. The controller is coupled to receive an event detected signal from the at least one event detector. The controller is further configured to control circumvention procedures of each peripheral based on the detected event and the level of susceptibility of the peripheral.

Abstract: A method, system and apparatus for server failure diagnosis and self-healing in a server farm. An autonomic server farm which has been configured in accordance with the inventive arrangements can include a multiplicity of servers enabled to respond to requests received from clients which are external to the server farm. A resource director such as an IP sprayer or a network dispatcher can be configured to route requests to selected ones of the servers, in accordance with resource allocation algorithms such as random, round-robin and least recently used. Significantly, unlike conventional server farms whose management of failure diagnosis and self-healing relies exclusively upon the capabilities of the resource director, in the present invention, client-assisted failure detection logic can be coupled to the resource director so as to provide client-assisted management of failure diagnosis and self-healing.

Abstract: The present invention provides for check-pointing an non-clustered workload to make room for a clustered workload that was running on a computer system that has suffered a hardware failure.

Abstract: A method for restoring a volume in a continuous data protection system begins by selecting a volume to restore. The selected volume is loaded into a workspace. The selected volume is copied from the workspace to a restore location. Operations are simultaneously performed on the selected volume while the selected volume is being copied. The selected volume is accessed on the restore location once the copying is complete.

Abstract: In a network system including a plurality of devices connected to a network, a master device supplies general devices with shared data to be shared by the devices in the system. On each update of the shared data, the master device generates backup data (the shared data before the backup) and successively transmits a backup instruction to all the devices managed thereby, starting from a device of the lowest priority in the system. Each of the devices successively receiving the backup instruction acquires the backup data from a device having an immediately higher priority than itself. By the above operation, reliable backup of the shared data (enabling restoration of the shared data to that of several times ago after the update is repeated several times) is realized without overloading any device in the system even when the storage capacity of each device for storing the backup data is relatively small.

Abstract: A method and apparatus for backing up data in response to detection of an imminent threat to the integrity of the data stored on the storage component a computing device is disclosed. The storage component may be a hard drive and the imminent threat may be a hard drive failure or a malware threat. In response to the receipt of an imminent threat, data stored on the storage component is copied to a computer-readable media either automatically or in response to user input. The backup procedure is configured by selecting data to backup and a media of storage on which to store the backup. Various sources of threat events are described.

Abstract: A global state management service manages replication in a distributed system. A distributed system can have several replica groups. The global state management service manages state associated with each replica group. Data replication is implemented without implementing an instance of a state machine on each device in a replica group. A replica group comprises a primary device and at least one secondary device. Each device in a replica group contains a replica of data of interest. The global state management service manages the allocation of primary and secondary devices. In the absence of a failure in either a primary device or a secondary device, read and write operations are performed without consulting the global state management service. When a failure is detected, the global state management service manages failover.

Abstract: A method is disclosed that enables accelerating the removal from service of a signal processor at a media gateway. In accordance with the illustrative embodiment of the present invention, the media gateway receives an indication to remove one of its digital signal processors from service. Instead of waiting for all of the calls that are using the signal processor to come to an end, the media gateway proactively moves the packet streams from the signal processor to one or more other signal processors resident at the gateway. Advantageously, an effort is made to avoid having to inform the corresponding media gateway controller of the move, thereby minimizing any discontinuity in each moved packet stream.

Abstract: A method, system, and storage medium for collecting bandwidth data is provided. The method includes producing master and slave text files in response to simultaneous collection of data samples from a network device by servers. The method also includes generating a clean data file by sorting data in the master and slave text files by the network device port, sorting data samples for the port by collection time, and for each of the samples: adding a designated interval of time to a time on the network device resulting in a target network device time whereby the time on the network device corresponds to a time the data sample was collected, examining data samples in the master and slave text files corresponding to the time the respective data samples were collected, selecting from one of the master and slave text files the sample with a collection time most closely matching the target network device time, and storing the selected sample in the clean data file.

Abstract: A computer network, system and computer-readable medium for completing a backup job that was interrupted during a backup process is described. The computer-readable medium causes a processor to perform the steps of retrieving an object from one or more volumes stored in a client computer, determining whether the object is listed in a catalog, where the catalog comprises a partially backed up volume from a list of one or more volumes that still need to be backed up, determining whether the object is partially backed up if the object is listed in the catalog and writing the data contained in the object to one or more storage devices if the object is partially backed up or is not listed in the catalog.

Abstract: An apparatus for deterministically performing active-active failover of redundant server blades hot-pluggable into a backplane of a network storage appliance chassis is disclosed. Each server monitors the other's heartbeat on a respective path in the backplane. Other paths between the two servers on the backplane enable one server to reliably kill the other server and take over its identity on the network in response to detecting a stopped heartbeat of the other server. The apparatus is superior to a conventional heartbeat link between servers in separate chassis, such as an Ethernet cable, because it is not prone to user removal or damage since the backplane cannot be removed by a user while the appliance is operational and enables each server to know a true heartbeat failure has occurred, as opposed to failure of a conventional external heartbeat link causing each server to each think the other has failed.

Abstract: A system comprises a plurality of processors, and data storage storing information that assigns a role of boot processor to one of the plurality of processors and assigns a role of spare processor to another of the plurality of processors. The system further comprises logic operable, responsive to detecting loss of lockstep for the boot processor, for transferring, during system runtime, the role of boot processor to the spare processor.

Abstract: To provide a multi-processing system capable of continuing an operation with efficiency even if a failure occurs in a processor during a system operation. A multi-processing system according to an embodiment of the invention includes: a memory storing a program and information necessary for executing the program; a processor manager reading the program from the memory, dividing the read program into threads with priority, and assigning the threads to at least two processors based on the priority of the threads and a processing state of each processor; the at least two processors executing the threads assigned by the processor manager; and a processing data manager storing information necessary for executing the thread assigned by the processor manager, and outputting the stored information necessary for executing the assigned thread to the memory if a failure occurs in one of the processors.

Abstract: A distributed architecture system including a plurality of nodes operatively coupled together by a bus. Each node includes a main controller configured to provide data to the bus and to an actuator, and a supervisory controller configured to provide data to the bus and to the actuator. Each node is configured such that during normal operations the main controller provides data to the actuator that controls the actuator and the supervisory controller generally does not provide data to the actuator that controls the actuator. Each node is configured such that if it is determined that the main controller is providing improper data, the supervisory controller provides data to the actuator that controls the actuator and the main controller does not provide data to the actuator that controls the actuator.

Abstract: The present invention discloses a method for preventing erroneous take-over in a redundant server system, and the method is used for a server system having a primary server for continuously producing heartbeat packets and sending the heartbeat packets to a redundant server of the server system. If the redundant server has not received any heartbeat packet, a redundant external transmission unit of the redundant server will send a detecting signal to the primary server. If the redundant external transmission unit has not received an acknowledged signal from the primary server, the redundant server will take over the primary server to continue providing network services.

Abstract: One embodiment disclosed relates to a method of compiling a program to be executed on a target central processing unit (CPU). The method includes opportunistically scheduling diagnostic testing of CPU registers. The method may include use of a predetermined level of aggressiveness for the scheduling of the register diagnostic testing. The scheduled diagnostic testing may include writing known data to a register, reading data from the register, and comparing the known data with the data that was read. If the comparison indicates a difference, then a jump may occur to a fault handler routine.

Abstract: This invention relates to a method for replaying, from a log file, events in a process belonging to a software application. This method in particular relates to internal events within a process belonging to an application executed in a multi-computer environment. This method comprises the following steps: reading (1) or receiving event data (KL) by a replay agent (PRE); from the event data, transmission (2) by the replay agent to a communication agent of message data addressed to said target process capable of initiating said event for this target process; transmission (3, 6) of said message data to the target process and initiation of said event.

Abstract: A fault-tolerant computer uses multiple commercial processors operating synchronously, i.e., in lock-step. In an exemplary embodiment, redundancy logic isolates the outputs of the processors from other computer components, so that the other components see only majority vote outputs of the processors. Processor resynchronization, initiated at predetermined time, milestones, and/or in response to processor faults, protects the computer from single event upsets. During resynchronization, processor state data is flushed and an instance of these data in accordance with processor majority vote is stored. Processor caches are flushed to update computer memory with more recent data stored in the caches. The caches are invalidated and disabled, and snooping is disabled. A controller is notified that snooping has been disabled. In response to the notification, the controller performs a hardware reset of the processors. The processors are loaded with the stored state data, and snooping and caches are enabled.

Abstract: Method and apparatus for transactional fault tolerance in a client-server system is described. In one example, output data generated by execution of a service on a primary server during a current epoch between a first checkpoint and a second checkpoint is buffered. A copy of an execution context of the primary server is established on a secondary server in response to the second checkpoint. The output data as buffered is released from the primary server in response to establishment of the copy of the execution context on the secondary server.

Abstract: Systems and methods for replacing software controlling active routers while minimizing impact on network operation are provided by virtue of one embodiment of the present invention. The software replacement process takes advantage of packet processor redundancy. An active packet switching device to be reprogrammed is de-activated and a redundant packet switching device takes over. The no longer active packet switching device is reprogrammed before being reactivated.

Abstract: Provided are a method, system, and program for remote copying of updates to primary and secondary storage locations subject to a copy relationship. A first copy relationship is established between first and second storage locations in a first storage system and second storage system, respectively, wherein updates to the first storage locations are copied to the second storage locations. A second copy relationship is established between the first storage locations in the first copy relationship and third storage locations in a third storage system, wherein updates to the first storage locations are indicated in a first bitmap. A suspended third copy relationship is established between the second storage locations in the first copy relationship and the third storage locations, wherein updates to the second storage locations are indicated in a second bitmap.

Abstract: A method for managing activity of a data storage system, including at least partly managing and performing an operator interaction with the storage system using a first processing system, and monitoring operation of the first processing system using a second processing system. The method further includes detecting a failure in operation of the first processing system using the second processing system and at least partly managing and performing the operator interaction using the second processing system in response to detecting the failure.

Abstract: An apparatus, system, and method are disclosed for updating firmware to a plurality of memory devices residing on disparate controllers wherein each controller includes an identical communication path address. A management module writes a firmware image to a memory device residing on a primary controller. The primary controller, in turn, copies the firmware image to another memory device residing on a secondary controller. In this manner, the apparatus, system, and method may update firmware on a plurality of memory devices. Because the primary controller is tasked with updating the secondary controller, the management module is only aware that the firmware of the primary controller has been updated. In this manner, updating of the firmware on the secondary controller is transparent to the management module.

Abstract: An information handling system recovers from memory errors associated with a memory unit that supports operation of an SMI handler by using another memory unit to support operation of the SMI handler. For example, if an SMI handler detects an error associated with a DIMM that supports operation of the SMI handler, then an SMI handler location module moves the SMI handler to another DIMM. For instance, a jump command is activated to jump to a pre-existing copy of the SMI handler stored at another DIMM. As another example, a relocation of the SMI handler to another DIMM is performed by changing address information used by the chipset and CPUs to run the SMI handler.

Abstract: A method of booting a multi-processor data processing device includes establishing a link between a first processor and a memory. The link is monitored to determine if, in response to a request from the processor, expected initialization data is communicated between the memory and the first processor. If unexpected data is detected on the link, the link is severed and a new link established between a second processor and the memory to allow the second processor to initiate the boot process. This ensures that, in the event of an error in the boot process at the first processor, the device can complete the boot process, thereby reducing device downtime.

Abstract: Data objects are sent using a data carousel and forward error correction. This involves segregating a file into groups, wherein each group represents k data blocks. From the k data blocks of each group, n erasure-encoded blocks are calculated, where n>k. The n erasure-encoded blocks are sent in a round-robin fashion using IP multicast technology: the first erasure-encoded block for each group, then the second block of each group, and so on. At a receiver, the blocks are stored on disk as they are received. However, they are segregated by group as they are stored. When reception is complete, each group is read into RAM, decoded, and written back to disk. In another embodiment, the receiver segregates allocated disk space into areas corresponding to sets of groups. Received blocks are then segregated only by set as they are written to disk. One or more RAM buffers can be used in this embodiment. When reception is complete, each set is read into RAM, decoded, and then written back to disk.

Abstract: Method and apparatus for facilitating device redundancy in a fault tolerant system is described. One aspect of the invention relates to common redundancy for a set of devices in a redundancy group. Each of the devices is in either an active role or a standby role. Virtual configurations for the devices are stored in a management system. Advertisements are periodically sent from each of the devices in the active role to each of the devices in the redundancy group. Each of the advertisements includes a configuration sequence number. An update in one of the virtual configurations is announced by incrementing the configuration sequence number in at least one of the advertisements. An updated virtual configuration is obtained at each of the devices in the standby role from the management system in response to detecting the configuration sequence number as incremented in the at least one advertisement.

Abstract: A method and system of copying memory from a source processor to a target processor by duplicating memory writes. At least some of the exemplary embodiments may be a method comprising stopping execution of a user program on a target processor (the target processor coupled to a first memory), continuing to execute a duplicate copy of the user program on a source processor (the source processor coupled to a second memory and generating writes to the second memory), duplicating memory writes of the source processor and duplicating writes by input/output adapters to create a stream of duplicate memory writes, and applying the duplicated memory writes to the first memory.

Abstract: A high availability, high performance software component architecture includes a primary configuration and at least one secondary configuration to take over from the primary configuration in the event of a failure thereof. The primary configuration includes at least one task processing component and a first client component and the secondary configuration includes at least one task processing component and one client component. Status data is present in both configurations and services for write access and read access for the task processing components are separated in such a way that status data read requests coming from the client components are divided between the configurations.

Abstract: A method of “stateful failover” is provided that allows email gateway systems in a cluster to deliver email messages that have been accepted for delivery by a member of the cluster, but has failed with out delivering the messages. The method involves creating a backup copy of the messages that have been accepted for delivery by one email gateway system in the stateful failover cluster on one or more other email gateway systems in the stateful failover cluster. Upon detecting the failure of the email gateway system that accepted the message, another member of the stateful failover cluster that has access to the backup copy of the message queue takes responsibility for the delivery of the messages on the mirrored queue.

Abstract: A computer device that includes a plurality of processor boards each provided with a processor, a memory, and a chipset, includes a first processor board that makes data in a cache, which have become unfixed as a result of an uncorrectable failure, invalid when the uncorrectable failure occurs on the first processor board in operation, and switches from the first processor board to a second processor board for replacement, and the second processor board that re-executes an instruction that was being executed in the first processor board when the failure occurred.

Abstract: A computer system including: a plurality of computers having at least one first computer and at least one second computer; a management computer adapted to manage the plurality of computers; and a storage system; the management computer including: a portion adapted to receive network identification information which is set to a network adapter of the first computer; and a portion adapted to transmit the network identification information of the first computer, to the second computer if the management computer receive a request to move a process executing on the first computer; the second computer including: a portion adapted to set the network identification information of the first computer to a network adapter of the second computer; a portion adapted to receive at least one program from a storage area of the storage system corresponding to the network identification information; and a portion adapted to boot the program.

Abstract: An apparatus and method of supporting the backup and recovery of a computing device. The computing device typically includes both a user computing environment and a supporting environment which enhances the stability and functionality of the user computing environment.

Abstract: Provided is a storage controller capable of changing a system configuration with scalability. A storage controller blade of the storage controller includes: interface units that each connect to one of a host computer and a disk device; a processor unit that controls a configuration of the storage controller and data access; a memory unit that stores a data cache and configuration information on the interface unit and the processor unit; and a mutual connection unit that connects the interface units, the processor unit, and the memory unit to one another. The processor unit recognizes a form of connection to another storage controller blade over the mutual network, and the connection path to another storage controller blade is set in the mutual connection unit based on the recognized form of connection.

Abstract: Method and system for performing a task in a control system is provided. The method includes, transmitting a command to a governing agent; re-transmitting the command from the governing agent to a primary, a secondary and an observer agent; validating the task information; and performing the task wherein, the task is executed by the primary agent while the secondary agent monitors task execution and if the primary agent fails to execute the task and/or a threshold value is exceeded, the secondary agent performs the task to the extent the secondary agent perceives that the task is not being performed and to the extent that it has sufficient capacity. The system includes an input module that transmits commands for executing a task to the governing agent that is operationally coupled to a processing unit.

Abstract: Described are techniques for performing data recovery processing for two failed devices included in a set of devices. It is determined whether the two failed devices are each one of the data devices, a horizontal parity device, or a diagonal parity device. Each of the data devices is partitioned into N?1 data segments. The horizontal parity device includes N horizontal parity segments each including parity information for a horizontal grouping of N?1 of the data segments. The diagonal parity device includes N diagonal parity segments each including parity information for a diagonal grouping of N?1 of the data segments. Recovery processing is performed to recover two failed devices in accordance with the diagonal parity device, the horizontal parity device, and the data devices.

Abstract: Provided are a method, system, and article of manufacture, wherein a primary control unit sends a selected time to a plurality of secondary control units. Clocks are periodically synchronized at the primary control unit and the secondary control units with a time server. The primary control unit and the plurality of secondary control units consistently update secondary storage subsystems coupled to the plurality of secondary control units at the selected time.

Abstract: A system that provides fault tolerance in a parallel processing system. During operation, the system executes a parallel computing application in parallel across a subset of computing nodes within the parallel processing system. During this process, the system monitors telemetry signals within the parallel processing system. The system analyzes the monitored telemetry signals to determine if the probability that the parallel processing system will fail is increasing. If so, the system increases the frequency at which the parallel computing application is checkpointed, wherein a checkpoint includes the state of the parallel computing application at each computing node within the parallel processing system.

Abstract: This invention relates to a method for logging events, in particular external, occurring in an application in a multi-computer environment, operating non-intrusively and transparently vis-à-vis said application. Said method comprises the following steps: reception (1) by the system software of a message addressed to said process and initiating for this process an external event; interception, by an interception agent, of message data representing said message or the event which it initiates; logging said event, by storing or sending (2) for storage logging data representing said message or the event which it initiates; transmission (4, 5) by the system software of said message data to said process.

Abstract: In accordance with one embodiment of the invention, a method for replicating data for redundant network components is disclosed. Data associated with a call is received and stored in a shared memory database. The data is sent to a mate network component of the network component if the network component is in an active mode. Sending of the data to the mate network component is terminated if the network component enters a mode other than the active mode.