Abstract: Node-centric checkpointing may be used in a multi-node computing system to provide fault-tolerance. Such checkpointing may involve storage of input and/or output data prior to and/or after execution of a task on a node.

Abstract: In a distributed computing system, a processing device executing a caveat enabled transaction manager registers one or more transaction participants with the transaction manager. To register a transaction participant, the caveat enabled transaction manager records an address of the transaction participant. The caveat enabled transaction manager additionally determines whether the transaction participant is associated with any transaction caveats. If the transaction participant is associated with a transaction caveat, the caveat enabled transaction manager records the transaction caveat.

Abstract: A method, apparatus, and system of a sequencing technique to account for a clock error in a storage area network are disclosed. In one embodiment, a system of a backup server includes a processing module to examine a data timestamped with a sequence of characters denoting a time according to a clock source, an analysis module to determine that the data has been timestamped at an earlier time than an other data previously received, a substitution module to provide the data an incremental sequence number placed with the data using an algorithm until a new data is received that includes a future timestamp with a later timestamp than the timestamp of the other data, and a storage module to store the data.

Abstract: A method for supporting compensation work comprises the steps of associating a data structure with a compensation function, and associating the data structure with the at least one event that can be encountered during execution of a forward work. The data structure contains a pair of pointers, which includes a first pointer to a code to be executed for the compensation function, and a second pointer to a stack frame of a closest lexically enclosing scope of a declaration of the compensation function in a computer program text.

Abstract: An application programming interface (API) that leverages operating system instrumentation to provide a chain of threads and processes may alleviate some debugging complications. Specifically, the chain may start with the first thread in the process that experienced the original failure and end with the last thread upon which the first thread directly or indirectly depends. The API may aid debugging efforts by classifying all threads related or dependent upon an original failed thread into specific categories of failures, requesting further information from the originating OS concerning specific failed threads, and using that information to debug the failed application or process more thoroughly.

Abstract: Mechanisms for aggressively optimizing computer code are provided. With these mechanisms, a compiler determines an optimization to apply to a portion of source code and determines if the optimization as applied to the portion of source code will result in unsafe optimized code that introduces a new source of exceptions being generated by the optimized code. In response to a determination that the optimization is an unsafe optimization, the compiler generates an aggressively compiled code version, in which the unsafe optimization is applied, and a conservatively compiled code version in which the unsafe optimization is not applied. The compiler stores both versions and provides them for execution. Mechanisms are provided for switching between these versions during execution in the event of a failure of the aggressively compiled code version. Moreover, predictive mechanisms are provided for predicting whether such a failure is likely.

Abstract: Techniques for generating a system model for use by and availability management framework (AMF) are described. Inputs are received, processed and mapped into outputs which are further processed into a configuration file in an Information Model Management (IMM) Service eXternal Markup Language (XML) format which can be used as a system model by an AMF.

Abstract: A method, apparatus, and program product restore database data having dependencies or ordering sensitivities, or that incorporates a database index. Unresolved dependencies are identified in response to an attempt to restore data in a computer. The data restoration is deferred and the data is cached when it has an unresolved dependency. After the dependency is resolved, the data restoration is completed. The unresolved dependency may be based-on data that must be restored or created prior to completing restoration. Deferred data is cached in memory and tracked in a deferred operation table. Data that is deferred may be accessed so long as operations that access the underlying based-on data are denied.

Abstract: A method, system and article of manufacture are disclosed for error recovery in a replicated state machine. A batch of inputs is input to the machine, and the machine uses a multitude of components for processing those inputs. Also, during this processing, one of said components generates an exception. The method comprises the steps of after the exception, rolling the state machine back to a defined point in the operation of the machine; preemptively failing said one of the components; re-executing the input batch in the state machine; and handling any failure, during the re-executing step, of the one of the components using a defined error handling procedure. The rolling, preemptively failing, re-executing and handling steps are repeated until the input batch runs to completion without generating any exception in any of the components that are not preemptively failed.

Abstract: Provided are techniques for static load balancing implemented in a filter driver. The filter driver determines a data quota for each of multiple data paths. The filter driver identifies a maximum number of commands for a target logical unit. The filter driver selects a data path on which to send a packet based on the data quota and the maximum number of commands. Also, provided are techniques for dynamic load balancing implemented in a filter driver. The filter driver determines a data transfer speed for each of multiple data paths. The filter driver updates a load balancing share for each of the multiple data paths based on the data transfer speed of each of the multiple data paths. The filter driver selects a data path on which to send a packet based on the load balancing share of each of the multiple data paths.

Abstract: A status of connectivity between servers of different sites (locations) is used to infer whether a network or a server failure has occurred such that data between the servers can be routed more efficiently reducing unnecessary network traffic due to duplicate messages. Servers may be grouped based on location or other characteristics and connectivity status determined based on the communication status of individual servers and their respective groups.

Abstract: A network device for processing packets. The network devices includes a ingress module for performing lookups for layer 2 switching and performing operations for maintaining a layer 2 table. When the ingress module updates the layer 2 table, the ingress module records the operation performed on the layer 2 table in a modification buffer. Entries are added to the modification buffer when the layer 2 table is modified and in the order in which the layer 2 table was modified. The network device thus enables reconstruction of the layer 2 table by performing the operations in the modification buffer.

Abstract: Fault tolerant operation is disclosed for a primary instance, such as a process, thread, application, processor, etc., using an active copy-cat instance, a.k.a. backup instance, that mirrors operations in the primary instance, but only after those operations have successfully completed in the primary instance. Fault tolerant logic monitors inputs and outputs of the primary instance and gates those inputs to the backup instance once a given input has been processed. The outputs of the backup instance are then compared with the outputs of the primary instance to ensure correct operation. The disclosed embodiments further relate to fault tolerant failover mechanism allowing the backup instance to take over for the primary instance in a fault situation wherein the primary and backup instances are loosely coupled, i.e. they need not be aware that they are operating in a fault tolerant environment.

Abstract: A system can test network performance of an electronic device via transmitting a testing file with a first designated name to a number of computers connected to the electronic device, obtaining comparison files from the number of computers after running the testing file, and replacing the names of the comparison files with standard names. The network performance of the electronic device can be confirmed via contents of the comparison files.

Abstract: There is provided a method of processing at least one data message using a computing device having at least a processor, a memory, a display device and an input device, the method including: retrieving from the memory and providing on the display device a number of operation icons each representing at least one operation to be performed on data; enabling a user of the computing device to select and arrange at least some of the operation icons on the display device using the input device; enabling the user to arrange graphical links between the selected operation icons to define the flow of data between the selected operation icons, wherein the selected operation icons and the graphical links together form a graphical representation of a process to be performed on the data message; enabling the user to specify at least one of the operation icons in the graphical representation as a transaction boundary node; defining at least one transaction region extending over part or all or the graphical representation

Abstract: Disclosed are a method, a system and a computer program product of operating a data processing system that can include or be coupled to multiple processor cores. In one or more embodiments, an error can be determined while two or more processor cores are processing a first group of two or more work items, and the error can be signaled to an application. The application can determine a state of progress of processing the two or more work items and at least one dependency from the state of progress. In one or more embodiments, a second group of two or more work items that are scheduled for processing can be unscheduled, in response to determining the error. In one or more embodiments, the application can process at least one work item that caused the error, and the second group of two or more work items can be rescheduled for processing.

Abstract: A method to enable an adaptive fault tolerance in a data system is provided. The method includes communicatively coupling a radiation hardened hardware element to at least one interrupt service routine application in a respective at least one processing element, triggering an interrupt service routine by an interrupt signal received from the radiation hardened hardware element at the interrupt service routine application after an execution of a code block by the at least one processing element, and inspecting the internal state of the at least one processing element to determine if an error occurred.

Abstract: A control logic device performs a local rollback in a parallel super computing system. The super computing system includes at least one cache memory device. The control logic device determines a local rollback interval. The control logic device runs at least one instruction in the local rollback interval. The control logic device evaluates whether an unrecoverable condition occurs while running the at least one instruction during the local rollback interval. The control logic device checks whether an error occurs during the local rollback. The control logic device restarts the local rollback interval if the error occurs and the unrecoverable condition does not occur during the local rollback interval.

Abstract: The present invention relates to a method and arrangements for node recovery particularly suitable for recovery involving large database fragments. The method for node recovery according to the present invention makes use of an old version of a fragment in a failed node (114). The old version is brought up to date by means of scanning an up to date version of the fragment in another node and copying parts of the fragment (117), which have changed since the creation of the old version, to the recovering node. A delete log is utilized in order to take deletions, which have taken place since the creation of the old version, into account in the recovery of the failed node (116). An advantage of the method for node recovery according to the present invention is that it does not require write transactions to be stopped during the recovery process.

Abstract: A read retry method performed in a hard disk drive, the read retry method may include performing a read operation; and ignoring a read error flag generated when a read error is generated, and continuing to perform the read operation.

Abstract: Various Embedded File System recovery techniques are described. A mobile computing device may comprise a radio, a processor coupled to the radio, and a memory coupled to the processor. The memory may be arranged to store a file system manager module (FSMM) that when executed by the processor manages an Embedded File System (EFS). The FSMM may create a factory image of the EFS on a periodic basis, or in response to a specific event such as a power down event. The FSMM may detect an error condition for the EFS, and replace the EFS with the factory image when an error is detected. Other embodiments are described and claimed.

Abstract: A storage controller changes a block size to carry out a shredding process. A data shredder uses a large block size BSZ1 set by a block size setting part to write shredding data in a storage area of a disk drive and erase data stored therein. An error arising during the writing operation of the shredding data is detected by an error detecting part. When the error is detected, the block size setting part sets the block size smaller by one stage than the initial block size to the data shredder. Every time the error arises, the block size used in the shredding process is diminished. Thus, the number of times of writings of the shredding data is reduced as much as possible to improve a processing speed and erase the data of a wide range as much as possible.

Abstract: A method and system for achieving highly available, fault-tolerant execution of components in a distributed computing system, without requiring the writer of these components to explicitly write code (such as entity beans or database transactions) to make component state persistent. It is achieved by converting the intrinsically non-deterministic behavior of the distributed system to a deterministic behavior, thus enabling state recovery to be achieved by advantageously efficient checkpoint-replay techniques. The method comprises: adapting the execution environment for enabling message communication amongst and between the components; automatically associating a deterministic timestamp in conjunction with a message to be communicated from a sender component to a receiver component during program execution, the timestamp representative of estimated time of arrival of the message at a receiver component.

Abstract: A system, method, and computer-accessible medium for logging and replaying asynchronous events are disclosed. One or more asynchronous events occurring during execution of a first instance of a computer program are logged. In logging the asynchronous events, a respective location in the execution of the first instance at which each of the one or more asynchronous events occurs is determined. A respective synchronous event preceding each asynchronous event is also determined. The asynchronous events are replayed during execution of a second instance of the computer program. In replaying each asynchronous event, the second instance is instrumented at the respective location during the execution of the second instance after detecting the preceding synchronous event.

Abstract: Techniques are provided for leveraging narrowband connectivity (such as dial-up communications or other types of low bandwidth communications) to provision or configure broadband connectivity between a broadband access provider and a broadband device, such as a DSL modem or a cable modem. Specifically, because narrowband connectivity does not require advance configuration or provisioning by the host system of connectivity parameters for an access-seeking device, a modem at an access-seeking device may be leveraged to establish a narrowband connection between that device and a host system and to enable an exchange or negotiation of connectivity parameters necessary to enable future broadband connectivity. Thus, once established, the narrowband connection may be used as a conduit for communicating required provisioning information between the broadband-enabling host and the access-seeking device to enable broadband connectivity by the device in the future.

Abstract: A logging last resource (LLR) system can provide a transaction log and transaction data to a LLR resource after a number of two-phase-commit resources have been prepared. The LLR resource manager can operate on the transaction log and transaction data in an atomic fashion so that the local commit can be done. The local commit can be done by the LLR manager in an atomic manner.

Abstract: An apparatus connected to a license management apparatus, storing license status data with license identifiers, via a network, and includes a part for receiving a request to acquire a license corresponding to a license identifier, a part for receiving designation of at least one electronic device for acquiring the license, a part for obtaining a license file corresponding to the license identifier from the license management apparatus, updating the license status data corresponding to the license identifier, and recording the license file to a storage part with the electronic device, a part for executing acquirement, and a part for executing re-execution in a case where the acquiring of the license by the electronic device is determined as a failure according to the result. The re-execution includes displaying the failure of the acquirement on a screen and re-executing the acquirement with the license file based on data input to the screen.

Abstract: Embodiments of the present invention provide techniques, including systems, methods, and computer readable medium, for dynamic atomic arrays. A dynamic atomic array is a data structure that provides an array that can grow or shrink in size as required. The dynamic atomic array is non-blocking, wait-free, and thread-safe. The dynamic atomic array may be used to provide arrays of any primitive data type as well as complex types, such as objects.

Abstract: Architecture that reduces data loss resulting from failover in an asynchronous log shipping deployment, but leveraging mid-tier and frontend servers to fill in lost data. In an asynchronous log shipping operation, a replication component asynchronously replicates messaging data to a backend server in accordance with one or more replication operations, which can be updates to databases on the backend server. These databases can include messaging data, such as email address books, mailboxes, etc. A history component maintains a history of replication operations on a frontend server. In the event of a lossy failover, a replay component is used for replaying the replication operations from the history to the backend server.

Abstract: Application-level replication, the synchronization of data updates within a cluster of application servers, may be provided by having application servers themselves synchronize all updates to multiple redundant databases, precluding the need for database-level replication. This may be accomplished by first sending a set of database modifications requested by the transaction to a first database. Then a message may be placed in one or more message queues, the message indicating the objects inserted, updated, or deleted in the transaction. Then a commit command may be sent to the first database. The set of database modifications and a commit command may then be sent to a second database. This allows for transparent synchronization of the databases and quick recovery from a database failure, while imposing little performance or network overhead.

Abstract: Provided are a method, system, and article of manufacture for synchronizing device error information among nodes. A first node performs an action with respect to a first node error counter for a device in communication with the first node and a second node. The first node transmits a message to the second node indicating the device and the action performed with respect to the first node error counter for the device. The second node performs the action indicated in the message with respect to a second node error counter for the device indicated in the message, wherein the second node error counter corresponds to the first node error counter for the device.

Abstract: A method for running a computer program on computing hardware, in particular on a microprocessor. The computer program includes multiple program objects designed as tasks, for example. Transient and permanent errors are detected during the running of the computer program on the computing hardware. To be able to handle these transient errors constructively when they occur in a computer system in such a way that the functionality and function reliability of the computer system are restored within the shortest possible error tolerance time, at least one program object that has already been sent for execution is set into a defined state on detection of an error and is restarted from this state. The program object is a runtime object of the computer program, for example, also known as a task. One or more tasks that are still being executed or have already been executed on occurrence of an error are restarted and run again.

Abstract: In an information processing apparatus for storing, based on a notification sent from a monitor unit used for monitoring a state of print processing of a printing apparatus, the status item presenting the state in a storage region, or deleting the status item stored in the storage region, and for displaying the status item stored in the storage region in a storage order, a monitor unit acquires a plurality of the status items stored in the storage region, compares priority of the acquired plurality of status items and priority of a new status item stored in the storage region, and notifies a control unit to store the status items in the storage region in an order based on priority obtained as a result of the comparison.

Abstract: A method for supporting compensation work comprises the steps of associating a data structure with a compensation function, and associating the data structure with the at least one event that can be encountered during execution of a forward work. The data structure contains a pair of pointers, which includes a first pointer to a code to be executed for the compensation function, and a second pointer to a stack frame of a closest lexically enclosing scope of a declaration of the compensation function in a computer program text.

Abstract: A method, apparatus, and program product utilize intelligent job functionality to diagnose an error in a computer. After detecting an error in a first job processing a task, and in response to another attempt to perform the task, a job selection algorithm selects a predetermined job in which to perform the task and diagnose the error. The predetermined job can be the first job or a new job associated with a signature that corresponds to the task that experienced the error. The predetermined job can be used to diagnose the error in a debugging session. Alternately, the first job may enter a debug session immediately after experiencing an error.

Abstract: Among other aspects disclosed are a method and system for processing a batch of input data in a fault tolerant manner. The method includes reading a batch of input data including a plurality of records from one or more data sources and passing the batch through a dataflow graph. The dataflow graph includes two or more nodes representing components connected by links representing flows of data between the components. At least one but fewer than all of the components includes a checkpoint process for an action performed for each of multiple units of work associated with one or more of the records. The checkpoint process includes opening a checkpoint buffer stored in non-volatile memory at the start of processing for the batch. For each unit of work from the batch, if a result from performing the action for the unit of work was previously saved in the checkpoint buffer, the saved result is used to complete processing of the unit of work without performing the action again.

Abstract: A storage controller changes a block size to carry out a shredding process. A data shredder uses a large block size set by a block size setting part to write shredding data in a storage area of a disk drive and erase data stored therein. An error arising during the writing operation of the shredding data is detected by an error detecting part. When the error is detected, the block size setting part sets the block size smaller by one stage than the initial block size to the data shredder. Every time the error arises, the block size used in the shredding process is diminished. Thus, the number of times of writings of the shredding data is reduced as much as possible to improve a processing speed and erase the data of a wide range as much as possible.

Abstract: The invention provides an apparatus, system, and method for cluster-wide peer recovery in the event of a computer failure. A failure of a first computer is detected and a recovery module is registered as the first computer. In one embodiment, the recovery module is a peer computer. The recovery module retrieves a privately held undo log data through the authorized assumption of the failure identity associated with the failed first computer, backs out in-flight transaction updates of the first computer, and frees up data resources locked by the first computer.

Abstract: A data scrubbing apparatus corrects disturb data errors occurring in an array of memory cells such as SMT MRAM cells. The data scrubbing apparatus receives an error indication that an error has occurred during a read operation of a grouping of memory cells within the array of memory cells. The data scrubbing apparatus may generate an address describing the location of the memory cells to be scrubbed. The data scrubbing apparatus then commands the array of memory cells to write back the corrected data. Based on a scrub threshold value, the data scrubbing apparatus writes the corrected data back after a specific number of errors. The data scrubbing apparatus may further suspend writing back during a writing of data. The data scrubbing apparatus provides a busy indicator externally during a write back of corrected data.

Abstract: One embodiment provides a system that facilitates the execution of a transaction for a program in a hardware-supported transactional memory system. During operation, the system records a failure state of the transaction during execution of the transaction using hardware transactional memory mechanisms. Next, the system detects a transaction failure associated with the transaction. Finally, the system provides an advice state associated with the recorded failure state to the program to facilitate a response to the transaction failure by the program.

Abstract: A method for transactional writing of data into a persistent memory comprising memory cells includes a transactional writing step and a transaction recovery step. The transactional writing step comprises one or more memory cell writing steps comprising the sub-steps of writing in a transaction buffer as transaction buffer entry the current data value and the corresponding address of the respective memory cell, writing a first valid marker for the memory cell in the transaction buffer, and writing a new data value to the memory cell. The transaction recovery step is performed in case of an abortion of the transactional writing step for restoring the current data values of the aborted transaction in the persistent memory, the transaction recovery step comprising the sub-step of writing a transaction recovery marker to the transaction buffer indicating the start of the transaction recovery.

Abstract: A computer system may receive an instruction to perform an action on an object. A schedule manager may identify a set of processes associated with the action that need to be performed by the computer system. A subset of primary process and a subset of secondary processes may be identified from the set of processes associated with the action. The schedule manager may schedule an asynchronous process to perform the subset of secondary processes on the object. The computer system may perform the subset of primary processes on the object. The subset of secondary process may be performed by the asynchronous process based upon the scheduling.

Abstract: Method and apparatus to provide bottleneck processing in a continuous data protection system having journaling. In an exemplary embodiment, system parameter information is collected and analyzed to output bottleneck information.

Abstract: Several methods and a system for analyzing ordered data using pattern matching over an indefinitely long ordered sequence of rows in a relational database are disclosed. In one embodiment, a method of a server includes receiving an ordered data in a relational database. The method further includes matching a pattern specified in a query on ordered data in a relational database in a single pass in constant space for overlapping mode of results. The method also includes creating an output data in the single pass in constant space for overlapping mode of results based on the matching of the ordered data with the pattern in the relational database query.

Abstract: A conductor transmits a commit processing request to directors in response to a commit request received from an application. Upon receipt of a response to the commit processing request from any one of the directors, the conductor notifies the application of a successful commit. If there is a pending transaction due to a failure in a director, the conductor instructs the director to process the pending transaction upon recovery from the failure. If a failure occurs in the conductor during transmission of the commit processing request and the commit processing request is not received by any of the directors, the conductor inquires the director whether there is a pending transaction after recovery from the failure, and instructs the director to process the pending transaction based on a processing result of the pending transaction obtained from another director.

Abstract: Apparatus and methods, such as those that read data from non-volatile integrated circuit memory devices, such as NAND flash. For example, disclosed techniques can be embodied in a device driver of an operating system. Errors are tracked during read operations. If sufficient errors are observed during read operations, the block is then retired when it is requested to be erased or a page of the block is to be written. One embodiment is a technique to recover data from uncorrectable errors. For example, a read mode can be changed to a more reliable read mode to attempt to recover data. One embodiment further returns data from the memory device regardless of whether the data was correctable by decoding of error correction code data or not.

Abstract: The self-healing system comprises a self-healing processor and an error mitigation system. The self-healing processor includes a code block associated with the operation of a portion of digital logic. The self-healing processor also includes a dynamic signature analysis circuit. The processor executes the code block. The dynamic signature analysis circuit creates a dynamic signature representing the operation of the portion of digital logic associated with the code block. The error mitigation system receives the dynamic signature from the dynamic signature analysis circuit. The error mitigation system compares the dynamic signature to a static signature to determine if the signatures match. If the signatures do not match, then the digital logic associated with the code block has an error. The error mitigation system retries execution of the code block. The error mitigation system stores log information describing the above events.

Abstract: The invention relates to computer engineering, and its usage ensures the possibility to recognize types of failures occurring during operation of the program computing means, and to respond respectively for a failure of each type. The failures can include errors in storing the core memory content, control transfer to wrong command errors, and errors relating to various time interruptions during a program run.

Abstract: Embodiments of circuits for processors with multiple redundancy techniques for mitigating radiation errors are described herein. Other embodiments and related methods and examples are also described herein.

Abstract: In accordance with embodiments, there are provided mechanisms and methods for performing compensation work. These mechanisms and methods can enable embodiments to provide design patterns for implementing compensation work. The ability of embodiments to provide design patterns for implementing compensation work can provide consistency of long-running business processes defined using means of some programming language or modeling tool. Further, embodiments can enable compensation functions implementing compensation work to be efficiently persisted and to be passed as arguments to other program components.