Abstract: A method to optimize the transmission of data from (N) primary backup appliances interconnected to a plurality of second backup appliances by a single communication link, wherein (N) is greater than 1, by transferring a data set to one or more secondary backup appliances by two or more of the (N) primary backup appliances using the communication link, and completing those transfers of the data sets by the two or more primary backup appliances at the same time.

Abstract: Disclosed is a technique for data synchronization. A first identifier for a portion of data on a primary volume is determined, wherein a unique identifier is associated with each portion of data at the primary volume. A second identifier for a portion of corresponding data at a secondary volume is determined, wherein a unique identifier is associated with each portion of data at the secondary volume. The first and second identifiers are compared. When the first and second identifiers do not match, the portion of corresponding data at the secondary volume in a storage device is replaced with the portion of data at the primary volume.

Abstract: An apparatus and method to replicate one or more files between non-symmetric storage systems are disclosed. The method supplies a first storage system comprising a first configuration, a first volume, a first file system, and a first replication appliance comprising first replication appliance memory. The method further supplies a second storage system comprising a second configuration, a second volume, a second file system, and a second replication appliance comprising second replication appliance memory, where the first configuration differs from the second configuration. The first storage system receives a dataset, writes that dataset to the first volume as a first file. Applicants' method then replicates the first file to the second volume as a second file, and maintains in the second replication appliance memory a second replication appliance mapping associating the first file with the second file.

Abstract: A method to adjust the data transfer rate for one of (N) primary backup appliances. The method forms by a first primary backup appliance at least one consistent transactions set. The first primary backup appliance receives the (n)th status signal, and the (n+1)th status signal from each of the other (N-1) primary backup appliances. The method calculates the (n)th effective bandwidth for each of the (N) primary backup appliances, the (n)th time to complete for each of the (N) primary backup appliances, and the (n)th effective aggregate bandwidth for all (N) primary appliances. If the (n)th time to complete for the first primary backup appliance is greater than the (n)th time to complete for each of the other (N-1) primary backup appliances, then the method provides at least one consistent transactions set from the first primary backup appliance to a first secondary backup appliance with no delay.

Abstract: Methods and apparatus are provided for copying data from a primary storage facility to a secondary storage facility which reduce the workload on the storage controller in the primary facility and minimize bandwidth usage. The primary storage facility includes a primary data replication appliance which transfers data to a secondary replication appliance. Updated data from a host is both stored through a storage controller in the primary facility and also received by the primary replication appliance. Logic in the primary replication appliance determines whether the immediately previous version of the data is in a buffer from a previous storage operation. If so, the current (updated) version of the data is compared with the previous version and the difference, such as calculated through a bit-wise exclusive-OR operation, is transferred to the secondary replication appliance.

Abstract: A method, apparatus, and article of manufacture containing instructions for processing multiple point-in-time consistent data sets. The method consists of creating multiple point-in-time data sets associated with a backup appliance which is associated with backup storage. Upon the transfer of a first update from a primary storage controller to the backup appliance, a determination is made whether the update affects data stored in the backup storage has been unchanged following the creation of the respective data sets. If the update does affect data which is unchanged following the creation of a data set, the current, non-updated data is read from the backup storage. A copy of the current, non-updated data thus read from the backup storage is stored in a storage location which can be one of the data sets or an independent memory location. In addition, pointers are associated with the remaining applicable data sets and directed to the copy at its storage location.

Abstract: An apparatus, system, and method are disclosed for reliably updating a data group in a data replication environment. The apparatus, system, and method reliably update the data group by receiving an updated data group sent from a first storage medium to a second storage medium, comparing the updated data group with a previous data group previously existing on the second storage medium and writing the updated data group to the second storage medium. The read-before-write and differencing method disclosed maintain reliability by storing multiple copies of changes made to the second storage medium during and after the write process.

Abstract: Provided are a method, system and program for prefetching data into cache. A prefetch command is processed that indicates at least one conditional statement and at least one block to prefetch from storage to cache in response to determining that the conditional statement is satisfied.

Abstract: A system, method and computer program product are provided to improve the efficiency of the transmission of consistency groups across multiple storage components by ensuring that volume transactions are evenly allocated among backup components. Each volume is initially assigned to a specified backup component. Once a color period begins, volume transactions are transmitted to the assigned backup component. Each backup component accumulates data transfer information for its assigned volumes throughout at least a portion of the color period and calculates a new volume assignment for the next color period. Before the start of the next color period, a captain storage controller transmits any new assignments to the backup components. During a next color period, the process is repeated using the new assignments.

Abstract: Disclosed is a technique for transferring data using multiple backup components. Responsibility for a portion of data is assigned to a first backup component. When a data update for the portion of data is received at the first backup component from a primary source, the data update is mirrored to a second backup component that is not assigned responsibility for the portion of data. Also disclosed is a technique for processing data updates with a group of backup components. It is determined that a new backup component is active. Near an end of a consistent transactions set formation period, responsibility for one or more portions of data are assigned to each backup component in the group and to the new backup component. During a next consistent transactions set formation period, data updates are processed with each backup component in the group and the new backup component. Moreover, disclosed is a technique for processing data updates with a group of backup components.

Abstract: A network component useful in tracking write activity by writing logs containing write address information is described. The tracking component may be used in networked systems employing data mirrors to record data block addresses written to a primary storage volume during the time a data mirror is unavailable. The tracking component can be available to any network originating node, and may therefore track write activity on multiple volumes. At the time a data mirror is reconstructed, the log written may be used to construct a list of block addresses pointing to locations on a primary storage volume wherein data differs from a secondary storage volume member of the mirror. The locations maybe copied from the primary to secondary storage volume to reconstruct the data mirror. The performance impact of the tracking component is minimal and a shared network resource is offered that increases fault tolerance in the event of backup device failures.

Abstract: Disclosed is a technique for data synchronization. A first identifier is determined for a portion of data at a first source. A second identifier is determined for a portion of corresponding data at a second source. The first and second identifiers are compared. When the first and second identifiers do not match, the portion of corresponding data at the second source is replaced with the portion of data at the first source.

Abstract: A method to adjust the data transfer rate for one of (N) primary backup appliances. The method forms by a first primary backup appliance at least one consistent transactions set. The first primary backup appliance receives the (n)th status signal, and the (n+1)th status signal from each of the other (N?1) primary backup appliances. The method calculates the (n)th effective bandwidth for each of the (N) primary backup appliances, the (n)th time to complete for each of the (N) primary backup appliances, and the (n)th effective aggregate bandwidth for all (N) primary appliances. If the (n)th time to complete for the first primary backup appliance is greater than the (n)th time to complete for each of the other (N?1) primary backup appliances, then the method provides at least one consistent transactions set from the first primary backup appliance to a first secondary backup appliance with no delay.

Abstract: A method to select a captain control node from a plurality of interconnected control nodes is disclosed. The method provides by each of the plurality of control nodes a first signal to each of the other control nodes, and then receives by each of the plurality of control nodes, a response signal from each of the other control nodes. The method then calculates by each of the plurality of control nodes individual response times for each of the other control nodes, and determines by each control node its aggregate response time. The method then determines whether to select a captain control node using the aggregate response times. If the captain control node is selected using the aggregate response times, then the method then determines the minimum aggregate response time, and designates the control node having that minimum aggregate response time the captain control node.