摘要:
The SCSI WRITE command is a two round trip protocol, which introduces significant latency in synchronous replication environments. Example embodiments of the present invention overcome these and other deficiencies by using SCSI READ commands from the replication site to the production site instead of using SCSI WRITE commands from the production site to the replication site to decrease latency in a synchronous replication environment. SCSI READ commands are single round trip commands. Thus, the number of round trips required to complete each I/O is reduced from two round trips to one round trip by maintaining at least one SCSI READ command from a SCSI initiator to a SCSI target and then responding to at least one of the at least one SCSI READ command at the SCSI target according to the SCSI READ command.
摘要:
In a first embodiment, computer implemented method and computer program product for executing an action selected from the group consisting of removing a LUN from a consistency group on a production site and adding a LUN to a consistency group on a production site; wherein the action is performed without loss of a journal tracking the changes to the consistency group at a replication site, the replication site replicating the consistency group on the production site.
摘要:
In one aspect, a method includes determining locations in a file that have changed using read signature commands, reading data around a changed file location for each file location that has changed, generating variable-size portions of the data read and sending the variable-size portions of the data to a backup system.
摘要:
A method, system, and program product for data deduplication in a replication environment, the replication environment having a production site, a splitter, and a replication site, wherein the replication site has a journal, comprising determining a digest for each chunk of data of a set of data chunks, determining for each chunk whether the digest is in an index on the production site, determining for which offsets are to be evicted from the cache on the replication site, replacing the chunks in set of chunks that are in the index with an offset, and transmitting the set of chunks, offsets and an eviction list to the replication site.
摘要:
In one aspect, a method includes providing data protection to data in a first volume at a first data protection appliance by storing a copy of the data in a second volume using a second data protection appliance, tracking changes between data locations in the first volume and the second volume using a delta marking stream (DMS) and receiving, at the first data protection appliance, metadata. The metadata is intended to be committed to the DMS. The method further includes mirroring the metadata at a third data protection appliance.
摘要:
Methods and apparatus for storage snapshots for logged access replication using a data protection appliance in a continuous data protection system having a splitter. In one embodiment, out of band data transfer reduces disruption on the replica site while accessing an image.
摘要:
A system, program product, and computer implemented method for data replication, comprising enabling a virtual service layer to consume a storage medium of a site and map the storage medium of the site as one or more virtual storage volumes, enabling the virtual service layer to present the one or more virtual storage volumes for consumption by a host, enabling a splitter of the site to intercept I/O written to the one or more virtual storage volumes, splitting intercepted I/O to a journal based replication appliance, and enabling the journal based replication appliance to create a continuous data protection image of the data written to one or more virtual storage volumes.
摘要:
A system, method, and computer product for data replication comprising receiving an IO, receiving an IO abort command, the IO abort command specifying an IO to be aborted, determining if the IO was sent to a DPA by a splitter.
摘要:
Methods and apparatus for a continuous data protection system having a protection agent and at least one data protection appliance that enables, in an image access mode, a rolling back of volumes in a highly available storage array on the target side so that the volumes contain data for a selected point in time, creation of a storage array snapshot of the volumes for the point in time, and user access to the storage array snapshot in the storage array.
摘要:
In an embodiment, a method and computer product is presented for executing a command in a replicated environment comprising a replication appliance and a production site, the method comprising: intercepting the command at a splitter; wherein the command comprises a atomic test and set request.