Abstract: The present invention is directed to a method of recovering a write ahead log after an interruption. In a first aspect of the present invention, a method of writing a log entry of a write ahead log may include initiating a log write to a write ahead log, the write ahead log having a first sector, and a second sector, wherein the first sector is followed by the second sector. A log entry including a sequence number is written to the second sector. Then, the log entry including the sequence number is written to the first sector.

Abstract: A processing system includes a local or local storage and a number of remote or remote storage systems that store data mirroring that maintained by the local storage system. Data that is written, changed, deleted or other wise modified by the local storage system is periodically sent to the remote storage systems via remote copy commands to update the mirroring data maintained by each. Failure of a local or a remote storage system will cause the surviving storage systems to exchange information indicative of the data maintained by each, and if differences exist, to exchange data so that the mirrored and/or mirroring data is the same.

Abstract: A data processing system having an efficient logging mechanism which stores log records for repairing a file system when its consistency is lost. When there is a transaction attempting to update metadata stored in metadata volumes, a metadata loading unit reads the requested metadata objects out of the volumes and loads them to a metadata cache. At that time, a metadata manager updates its internal database to record from which metadata volume each metadata object has been fetched. Each time the transaction updates a metadata object in the cache, a log collection unit collects a copy of the updated metadata object, together with a volume ID which indicates its home metadata volume. The collected data is temporarily stored in a log buffer, and finally saved into a log volume by a log writing unit.

Abstract: A method and apparatus for merging log data for recovering one or more database data sets to any point-in-time in a single pass of log data sets. Using disaster recovery technology, the invention may be implemented to provide a method to reduce the time that a broken database is unavailable by processing recovery input data in parallel and recovering multiple database data sets, including areas, simultaneously. The invention reduces operational complexity and resource constraints experienced by IMS customers trying to recover database data sets especially in a data sharing environment. The invention simplifies the recovery process by eliminating both the need for change accumulation and the need to run separate recovery jobs for each database data set requiring recovery, thereby reducing the impact of recovery on system resources.

Abstract: This invention relates to a method of generating a test-instruction string to test the pipeline mechanism of a processor, which automatically generates from randomly generated instructions an instruction string which causes a pipeline interlock. This invention comprises a table for notifying the subsequent instruction of the status of resource usage of the leading instruction and, by generating the resources used, by the subsequent instruction according to the status of resource usage of the table automatically generates a subsequent instruction that interferes with the leading instruction.

Abstract: A method and system for implementing failover in a computer system is provided. For a failover involving an order-sensitive dataset or application, a mechanism is provided to detect order inconsistencies between a first set of data and a second set of data, in which the first set of data corresponds to a first connection established before the failover and the second set of data corresponds to a second connection established after the failover.

Abstract: A journaling method is provided for supporting a recovery when a system is abnormally terminated in a shared disk environment. When a system call operation to take part in a journaling is generated, in order to guarantee a recovery, a transaction is started and new transaction region is assigned. Then, a system is initialized and a transaction type is set up. Lock information on modified data is acquired and added to the transaction so that a transaction manages lock information. A reflection to a disk during a modification of metadata is prevented. Modified metadata added to the transaction and modified information on principal general data are recorded. Then, lock information connected to the transaction is released.

Abstract: A method and system for improving data migration from source data stripes to destination stripes in a Redundant Array of Independent Drives (RAID) logical drive migration (LDM). The invention describes a procedure for checkpointing data migrations, so that if the system should fail in the middle of the LDM, the computer can resume the LDM from the last checkpointed destination stripes, thus ensuring accurate tracking of data location. Further, the invention also provides the capability of properly checkpointing a data write to a stripe according to whether or not the stripe unit has been migrated previously.

Abstract: A storage device for a computer, combining an audit trail storage with (1) a connector adapted for connecting the storage device to a connection point of the computer, (2) a primary storage having first data stored thereon, and adapted for sending the first data to the computer in response to a corresponding read command received at the connection point, and (3) secondary storage adapted for storing second data in response to a corresponding write command received at the connection point and further adapted for sending the second data to the computer in response to a corresponding read command received at the connection point. The audit trail storage is adapted for recording a log of communication activity occurring at the connection point.

Abstract: Described is an approach for recovering the failure of a transaction. According to the approach, a first change and a third change is made to a first resource and a second change is made to a second resource. The second change was made after the first but before the third. However, to recover the failure of the transaction, a recovery record for the third change is applied before the recovery record for the second change. Also described is an approach involving establishing links that link together a set of undo records that contain changes made to the particular resource. Also described is an approach for applying two or more undo records in parallel.

Abstract: A method, program product and apparatus for checkpointing for recovery of channels in a data processing system using a protocol which allows for multiplexing operations at the frame level and streaming of commands and data. For unsuccessful retries, the correct primary CCW address is reported back to software indicating the extent to which the channel completed modifying and accessing S/390 storage.

Abstract: Provided are a method and apparatus for providing serialization support for computer systems and computer system resources, in particular for application programs which require sequencing and/or consistency of data. The computer system resource includes serialized processes with units of work. A serialization token is provided at the start of a serialized process including at least one unit of work. The serialization token remains in force until the or all of the units of work of the process have been committed or rolled back. The serialization token controls the commencement of other processes of the computer system during the time that it is in force. The serialization is carried out by storing an abstract representation of the process to be serialized in storage which is accessible by all the programs and computers where the serialization applies.

Abstract: An apparatus, program product and method manage a queue through the use of element-based journaling to record changes made to logical elements in a queue. In contrast to convention memory-based journaling, where any changes to the memory representing an element in a queue are journaled, only those operations that affect the logical ordering and/or placement of an element on a queue, or the logical contents of such an element, are journaled. Memory management operations such as pointer manipulation operations that modify pointers or other memory management data, but do not otherwise modify the actual elements in a queue or their relative ordering, need not be journaled. As a consequence, the storage and processing overhead associated with journaling may be substantially reduced, thereby substantially reducing the overhead associated with maintenance and recovery of a queue.

Abstract: A data processing system and method that provides two processes, checkpointing and compute point propagation, and permits a continuous flow of data processing by allowing each process to (1) return to normal operation after checkpointing or (2) respond to receipt of a compute point indicator, independently of the time required by other processes for similar responsive actions. Checkpointing makes use of a command message from a checkpoint processor that sequentially propagates through a process stage from data sources through processes to data sinks, triggering each process to checkpoint its state and then pass on a checkpointing message to connected “downstream” processes. A compute point indicator marks blocks of records that should be processed as a group within each process. A compute point indicator is triggered and sequentially propagates through a process stage from data sources through processes to data sinks without external control.

Abstract: A method and apparatus for removing changes made by a dead transaction is provided. According to the method, a first change is performed by the dead transaction prior to a second change. The first change made by the dead transaction is then undone prior to undoing the second change made by the dead transaction. According to another aspect of the invention, a method and apparatus for applying changes in redo records to make a particular resource reflect changes made to the particular resource in volatile memory before a failure is provided. The method includes establishing links that link together a set of redo records that contain changes made to the particular resource. The links are then followed to apply the changes contained in the set of redo records to cause the particular resource to reflect the changes made to the particular resource in volatile memory before the failure.

Abstract: Utilizing only a single input/output (I/O) session to commit a single message transaction to a non-volatile storage within a message transaction system is disclosed. In one embodiment, a method first determines an error detection value, such as a checksum, for a single transaction. The single transaction and the value are written to the non-volatile storage in only a single I/O session. The single transaction is committed upon the successful writing of the transaction and the error detection value to the storage. Upon recovery from a system failure, the error detection value for the transaction is redetermined; if it does not match the value as has been stored to the storage, then the transaction is aborted. Otherwise, the transaction has been committed.

Abstract: An adaptive forward error correction technique based on noise bursts and the rate at which they occur is disclosed. The forward error correction parameters are determined using statistics describing the noise burst duration and period. The occurrence, duration and period of the noise burst are determined by the error vector magnitude calculated during the decoding process.

Abstract: A protection data control unit is provided for a cache control unit. In the writing operation to write user data from an upper apparatus into a cache memory, protection data is formed on a data block unit basis of the user data and written into the cache memory. When the interruption of the transfer of the data block is detected during the writing into the cache memory, a forming state of the protection data upon interruption is stored in the cache memory. When the transfer restart of the data block is detected after the interruption, the forming state is returned to the forming state of the protection data upon interruption stored in the cache memory and the formation of the protection data is restarted.

Abstract: A pipelined processor and method are disclosed including an improved history file unit. The pipelined processor processes a plurality of instructions in order. A register file is included which includes a different read port coupled to each register field in an instruction buffer for reading data from the register file. A history file unit is included and is coupled to each of the read ports of the register file for receiving a copy of all data read from the register file.

Abstract: A data processing system and method that provides checkpointing and permits a continuous flow of data processing by allowing each process to return to operation after checkpointing, independently of the time required by other processes to checkpoint their state. Checkpointing in accordance with the invention makes use of a command message from a checkpoint processor that sequentially propagates through a process stage from data sources through processes to data sinks, triggering each process to checkpoint its state and then pass on a checkpointing message to connected “downstream” processes. This approach provides checkpointing and permits a continuous flow of data processing by allowing each process to return to normal operation after checkpointing, independently of the time required by other processes to checkpoint their state.

Abstract: A method and mechanism is described for reliably transmitting messages over an unreliable transport mechanism. A sender transmits a first message over an unreliable mechanism to a receiver, and a second message that identifies the first is transported over a reliable transport mechanism to the receiver. When the receiver receives the second message, the receiver determines whether it has received the first message. If not, the receiver requests retransmission of the first message.

Abstract: An information delivery system includes receiving platforms to receive data communicated from a transmitting source according to a unidirectional transfer protocol over a link. At least some receiving platforms process the received data to determine if a portion of expected data is missing. If so, a receiving platform is able to generate and send a request for the missing data according to a supplemental protocol that is separate from the unidirectional transfer protocol. The transmitting source is able to receive the request for missing data and to retransmit the missing data. The data may include ancillary information (e.g., enhancement data) associated with audio/video content (e.g., television content). By adding the supplemental protocol to enable requests for missing data, a unidirectional transfer protocol that is not reliable is made reliable as a mechanism is provided to ensure receipt of the transmitted data.

Abstract: A method for communicating transactions includes providing an interconnect having a plurality of ports for communicating transactions between a plurality of domains in a computing system is provided. Each port is associated with a subset of the domains. The interconnect includes a first signal path for transmitting a first portion of the transaction and a second signal path for transmitting a second portion of the transaction. A transaction issued from a port associated with more than one of the domains is identified. An error in one of the first and second portions of the transaction is identified. The transaction is canceled responsive to identifying the error. A computing system for communicating transactions includes first and second devices. The first device is adapted to receive a first portion of a transaction. The second device is adapted to receive a second portion of the transaction in lockstep with respect to the first device.

Abstract: A primary computer system has a database, application programs that modify the local database, and a transaction manager that stores audit records in a local image trail reflecting those application program modifications to the local database. In a remote backup system, a Receiver process receives audit records from the primary system. The audit records include audit update and audit backout records indicating database updates and database backouts generated by transactions executing on the primary system. The Receiver stores the audit update and audit backout records in one or more image trails. For each image trail there is an Updater process that applies to a backup database volume the database updates and backouts indicated by the audit update and audit backout records in the image trail. The remote backup system periodically executes a file purge procedure, which identifies the oldest transaction table from among the transaction tables in the last image trail file accessed for each of the image trails.

Abstract: A parallel logging and recovery scheme for highly available main-memory database systems is presented. A preferred embodiment called parallel redo-only logging (“PROL”) combines physical logging and selective replay of redo-only log records. During physical logging, log records are generated with an update sequence number representing the sequence of database update. The log records are replayed selectively during recovery based on the update sequence number. Since the order of replaying log records doesn't matter in physical logging, PROL makes parallel operations possible. Since the physical logging does not depend on the state of the object to which the log records are applied, the present invention also makes it easy to construct a log-based hot standby system.

Abstract: Methods are provided for preventing data from being written to a data sector that has been identified as faulty on a disk in a disk drive system. It is first determined whether data of a data sector of a portion of a disk is recoverable. If the data is non-recoverable, an invalid logical ID is written to the logical block address (LBA) corresponding to the data sector thereby identifying the data sector as faulty and preventing data from being written at the data sector. The LBA is then reallocated to point to a different data sector if the data is non-recoverable. Thus, further data directed to the LBA will be written to the different data sector responsive to the reallocated LBA, thereby avoiding writing the new data to the data sector identified as faulty. Internal drive read re-tries can still take place on the faulty data sector even after the LBA has been assigned an invalid logical ID.

Abstract: Audit Trail recovery is enhanced by including addresses of immediately prior periodic saves of all active or open Steps or transactions in each new period save area. Reduced cost in main memory usage and on-the-fly processing to accomplish audit trail format that enhances recovery time results from ability to immediately address next prior period save data once a last periodic save is found in reading back the audit trail. Reading back the audit trail to find the last periodic save can be avoided too by directly addressing the last periodic save in preferred embodiments since the computer system will in all events temporarily maintain a record of the address of the last periodic save to record the next time a periodic save is to be made, so this temporarily maintained record can be accessed and used to jump directly to the last periodic save address on the audit trail.

Abstract: When a processor becomes inactive, queues resident in the memory of that processor become inaccessible. Thus, in order to access those queues, an active processor takes over ownership of one or more of the inaccessible queues. Each queue may be rebuilt prior to its take over to ensure that it has up-to-date information.

Abstract: A “Selective Undo Function” for computer programs allows a user to select any single specific action that was previously recorded by the computer, and undo only that selected action, rather than every action that chronologically follows the specific action. Specifically, the computer program may undo just the selected action, even if the selected action is not the last action taken by the user, if that is possible; or the computer program may perform some analysis and undo other actions that are deemed prerequisites to undoing the selected action, upon user confirmation; or if the analysis performed by the computer program cannot determine what these prerequisites are or cannot perform the undo function, then the computer program may take a default action, e.g., undo the selected action as well as all actions that follow the selected action, in chronological order.

Abstract: An independent checkpointing method using a memory checkpoint on a distributed system that includes a message transmission routine, a message processing routine, and a periodical checkpoint routine. The message transmission routine adds a self checkpoint number to a message to be transmitted when a current process tries to send a message to another process. The message processing routine performs a memory checkpoint and processes a message in reference to a checkpoint number of a transmission process, a checkpoint number of the current process, a memory checkpoint flag, and a message transmission flag when a message is received from a process. The periodical checkpoint routine performs a checkpoint that records a necessary state information for recovery against faults periodically in reference to the memory checkpoint flag.

Abstract: A means for guaranteeing the proper behavior as specified by the JMS semantics of clustered message server when the individual computer that comprise the cluster are separated by a network partition. A clustered message server is responsible for the reliable transportation of messages between different distributed computer applications. It employs multiple computers to perform a function that otherwise appears to be performed by a monolithic server running on one computer, but with more capacity and reliability than can be provided by one computer. If a computer in the cluster fails, another computer should automatically assume the role of the failed computer. However, it is not possible for the other machines in the cluster to detect the difference between the failure of one or more computers in the cluster, and the failure of data network connecting those computers.

Abstract: The method of the present invention comprises the steps of: provisionally updating a resource through a first transaction program and locking the resource; determining if a second transaction program generates a request to update the resource which has been locked because of the provisional update through the first transaction program; making the second transaction program exclusively wait and reproducing a third transaction program having an internal status identical to that of the second transaction program when the second transaction program generates the update request; providing a virtual resource having an original status before the provisional update through the first transaction program; provisionally updating the virtual resource through the third transaction program without exclusive wait; and rolling back and closing the exclusively waiting second transaction program and executing commit in the reproduced third transaction program when rollback of the first transaction program is executed due to an

Abstract: A system and method are disclosed for the automatic and transparent repair of IP multicast sessions. The invention is a system and method for the repair of IP multicast sessions. In one aspect of the invention the method repairs a multicast session in a network, beginning with the step of sending a request message from a source to a subscription server in the network, requesting a repair service for an original multicast session originated by the source. The method continues by sending an enabling signal from the subscription server to a plurality of retransmit servers in the network, to buffer data traffic from the original multicast session, in response to the request. The method continues by buffering a copy of the data traffic at each of the plurality of retransmit servers and monitoring errors in each copy. The method continues by automatically selecting with the plurality of retransmit servers at least one retransmit server from among the plurality, having a minimum of the errors in its respective copy.

Abstract: A fault-tolerant transaction processing system and method stores records associated with operations of the system in order to permit recovery in the event of a need to roll back a transaction or to restart the system. At least some of the operational records are stored as a recovery log in low-speed non-volatile storage and at least some are stored as a recovery list in high speed volatile storage. Rollback of an individual transaction is effected by reference to the recovery list whereas restart of the system is effected by reference to the recovery log.

Abstract: A method and system for recovering after failure of a first server to which a client was connected for accessing a resource is provided. When it is detected that a connection between the first server and the client has failed, the client is automatically connected with a second server that has access to the resource. After automatically connecting the client, the client accesses the resource through the second server. The client stores information about the state of the session with the first server so that processing can continue where it left off after the client connects with the second server. The client may be pre-connected to the second server prior to the failure of the first server to reduce the latency caused by switching in response to a failure. The second server may be configured to pre-parse the commands that the client issues to the first server to further reduce the latency associated with switching to the second server.

Abstract: Disclosed is a method, system, and article of manufacture for processing modified meta data for data recovery operations. The meta data provides information on user data maintained in a storage device. The system determines whether meta data tracks maintained in a cache were modified and indicates in a non-volatile memory that the determined meta data tracks were modified. Data recovery operations may be initiated as a result of a system failure, such as a warmstart or coldstart recovery. During such data recovery operations, the system processes the non-volatile memory and the indications of modified meta data tracks therein to rebuild lost meta data tracks in the cache.

Abstract: The technique disclosed herein provides for simultaneously checkpointing all of the processes in a specified process group or family at the application level, and restoring those processes at a later time, optionally on a different machine, with the parent-child relationships restored intact. File descriptors that are shared among the processes at checkpoint time are restored to the family of processes such that the file descriptors are shared just as they were at the time of taking the checkpoint.

Abstract: A checkpoint of a parallel program is taken in order to provide a consistent state of the program in the event the program is to be restarted. Each process of the parallel program is responsible for taking its own checkpoint, however, the timing of when the checkpoint is to be taken by each process is the responsibility of a coordinating process. During the checkpointing, various data is written to a checkpoint file. This data includes, for instance, in-transit message data, a data section, file offsets, signal state, executable information, stack contents and register contents. The checkpoint file can be stored either in local or global storage. When it is stored in global storage, migration of the program is facilitated. When a parallel program is to be restarted, each process of the program initiates its own restart. The restart logic restores the process to the state at which the checkpoint was taken.

Abstract: Provided are a method and apparatus for providing serialization support for computer systems and computer system resources, in particular for application programs which require sequencing and/or consistency of data. The computer system resource includes serialized processes with units of work. A serialization token is provided at the start of a serialized process including at least one unit of work. The serialization token remains in force until the or all of the units of work of the process have been committed or rolled back. The serialization token controls the commencement of other processes of the computer system during the time that it is in force. The serialization is carried out by storing an abstract representation of the process to be serialized in storage which is accessible by all the programs and computers where the serialization applies.

Abstract: According to an approach for propagating messages in a distributed computer system, message information is identified that needs to be sent to a destination site. After the message information is identified, the message information is assigned a propagation sequence number that identifies when the message information is sent to the destination site relative to other message information sent from the source site to the destination site. A message that is based on the message information is then transmitted to the destination site. The transmitted message includes the sequence number value and a source ID that identifies the source site as transmitting the message to the destination site. After the message is received at the destination site, the propagation sequence number that was assigned to the message information may be stored in nonvolatile memory at the destination site.

Abstract: A “Selective Undo Function” for computer programs allows a user to select any single specific action that was previously recorded by the computer, and undo only that selected action, rather than every action that chronologically follows the specific action. Specifically, the computer program may undo just the selected action, even if the selected action is not the last action taken by the user, if that is possible; or the computer program may perform some analysis and undo other actions that are deemed prerequisites to undoing the selected action, upon user confirmation; or if the analysis performed by the computer program cannot determine what these prerequisites are or cannot perform the undo function, then the computer program may take a default action, e.g., undo the selected action as well as all actions that follow the selected action, in chronological order.

Abstract: A method, apparatus, and article of manufacture for performing a two-phase commit protocol for a transaction in a system having a plurality of data sources. An availability status is verified for all of the data sources, and the two-phase commit protocol for the transaction is completed for those data sources that are available, while the transaction is logged for data sources that are unavailable.

Abstract: The present invention describes a method and system for providing cluster replicated checkpoint services. In particular, the method provides cluster replicated checkpoint services for replicas of a checkpoint in a cluster. The cluster includes a first node and a second node, which are connected to one another via a network. The replicas include a primary replica and a secondary replica. The method includes managing the checkpoint that contains checkpoint information, and creating the primary replica in a memory of the first node. The primary replica contains first checkpoint information. The method also includes updating the primary replica so that the first checkpoint information corresponds to the checkpoint information, creating the secondary replica that contains second checkpoint information in a memory of the second node, and updating the secondary replica so that the second checkpoint information corresponds to the checkpoint information.

Abstract: A replica system and a method of producing a replica are provided for dealing with a database management system including both database programs and files. The system transmits and receives data between interfaces using a common data format, by including a replica manager which receives a replica instruction and determines an object to be read from a master database and an object to be written in a copy database, according to the replica instruction. By giving a command to the database management system, data associated with the object to be read is read from the database, and data format is converted into a standard data format. The received data is then converted to data format associated with the object to be written, and the converted data is finally stored in the copy database.

Abstract: A message queuing system is provided that saves and stores messages and their state in an efficient single file on a single disk to enable rapid recovery from server failures. The single disk, single file storage system into which messages and their states are stored eliminates writes to three different disks, the data disk, the index structure disk and the log disk. The single disk, single file storage is made possible by clustering all information together in a contiguous space on the same disk. The result is that all writes are contained in one sweeping motion of the write head in which the write head moves only in one direction and only once to find the area where it needs to start writing messages and their states are stored.

Abstract: A method and system for reducing overhead associated with recovering after a failure. According to the method, a checkpoint value is maintained that indicates which records of a plurality of records have to be processed after the failure. The plurality of records contain change information that corresponds to a plurality of data blocks. A target checkpoint value is determined based on a desired number of data block reads that will be required during a redo phase of recovery. Changes contained in volatile memory are then written to nonvolatile memory to advance the checkpoint value to at least the target checkpoint value. According to another aspect of the invention, the target checkpoint value is determined using a circular queue of offset buckets. The offset buckets are used to store index values that are associated with buffers in the ordered list. The target checkpoint value is periodically set equal to an index value that is contained in an offset bucket.

Abstract: Disclosed is a device adapter for controlling devices in a network comprising computer processor nodes and one or more devices, the device adapter having means for determining whether or not a device state has changed after a failure. Responsive to a determination that the device state has not changed, the adapter communicates with other device adapters in the network to reestablish permissions before resubmitting I/O requests. Responsive to a determination that the device state has changed after a failure, the adapter reintegrates itself with the other device adapters in the network before reprocessing work as necessary.

Abstract: A complete and consistent set of checkpoint files is captured and identified for use in restarting a parallel program. When each process of a parallel program takes a checkpoint, it creates a checkpoint file. The checkpoint file is named, and a part of that name includes a version number for the checkpoint file. When a parallel program is to be restarted, each process of the parallel program identifies its most current valid checkpoint file. It provides the version number of this file to a coordinating process. The coordinating process then decides which version of the checkpoint files is valid and complete for all of the processes of the parallel program participating in restart. Once this version number is determined, it is forwarded to the processes and the processes restore themselves using its corresponding checkpoint file having that particular version number.

Abstract: A data processing system and method that provides two processes, checkpointing and compute point propagation, and permits a continuous flow of data processing by allowing each process to (1) return to normal operation after checkpointing or (2) respond to receipt of a compute point indicator, independently of the time required by other processes for similar responsive actions. Checkpointing makes use of a command message from a checkpoint processor that sequentially propagates through a process stage from data sources through processes to data sinks, triggering each process to checkpoint its state and then pass on a checkpointing message to connected “downstream” processes. A compute point indicator marks blocks of records that should be processed as a group within each process. A compute point indicator is triggered and sequentially propagates through a process stage from data sources through processes to data sinks without external control.

Abstract: A method of delivering highly-reliable, fault-tolerant communications services in a telecommunications network of distributed call processing systems. The method advantageously identifies a set of objects within the telecommunications network requiring checkpointing; checkpoints the objects; and subsequently restores the checkpointed objects in the event of a failure. Various aspects of the method are disclosed, including restoration strategies.