Abstract: Systems and methods for providing any point in time functionality. With a storage system such as a VSAN, any point in time protection is achieved by combining a metadata stream with snapshots of the storage system. This allows snapshots to be generated in hindsight such that any point in time functionality is provided.

Abstract: Restoring data stored by a backup process in an object store of a cloud network by initiating a restore request to a server accessing the object store, by providing a source identifier of one or more data blocks of the requested data and a timestamp for a backup image at a requested point in time (PIT). The server manages a queue of blocks in temporary storage to provide blocks for retrieval by the client, and initially fills the queue with blocks from a full backup and unchanged by any incremental backup to the requested PIT. In response to the request, the server assembles blocks that are changed from the full backup through one or more incremental backups to the requested PIT using a full block synthesis process employing forward and reverse delta objects, and sends to the client the blocks initially filled in the queue with the assembled blocks to return a full PIT backup image for the requested PIT.

Abstract: Embodiments for optimizing a cost of storing backup data in an object store of a public cloud network provided by a cloud provider by storing delta objects each embodying changes to blocks of the backup data between a full backup incremental backups. The delta objects are each of a same block size, which can be changed. The process selects two sequential delta objects to form a larger object by extending each block to a minimal part size that is larger than the block size and inserting a region of unchanged data between the two delta objects. The larger object is divided into three different parts comprising the inserted region, and each of the extended delta object blocks. The process calculates an API and a compute cost for a synthesis operation on the delta objects based on a cost model and current price schedule of the cloud provider for the three different parts, and for the substantially larger object as a whole. The process selects a block size for the delta objects based on a lowest calculated cost.

Abstract: Restoring data stored by a backup process in an object store of a cloud network by initiating a restore request to a server accessing the object store, by providing a source identifier of one or more data blocks of the requested data and a timestamp for a backup image at a requested point in time (PIT). The server manages a queue of blocks in temporary storage to provide blocks for retrieval by the client, and initially fills the queue with blocks from a full backup and unchanged by any incremental backup to the requested PIT. In response to the request, the server assembles blocks that are changed from the full backup through one or more incremental backups to the requested PIT using a full block synthesis process employing forward and reverse delta objects, and sends to the client the blocks initially filled in the queue with the assembled blocks to return a full PIT backup image for the requested PIT.

Abstract: Embodiments for optimizing a cost of storing backup data in an object store of a public cloud network provided by a cloud provider by storing delta objects each embodying changes to blocks of the backup data between a full backup incremental backups. The delta objects are each of a same block size, which can be changed. The process selects two sequential delta objects to form a larger object by extending each block to a minimal part size that is larger than the block size and inserting a region of unchanged data between the two delta objects. The larger object is divided into three different parts comprising the inserted region, and each of the extended delta object blocks. The process calculates an API and a compute cost for a synthesis operation on the delta objects based on a cost model and current price schedule of the cloud provider for the three different parts, and for the substantially larger object as a whole. The process selects a block size for the delta objects based on a lowest calculated cost.

Abstract: Storing data in an object store in cloud-based storage for data protection applications to address key user workflows, and provide an easily and seamlessly scalable system that minimizes the cost economics of the various resources, considers how backups are transmitted and which target data is most important during a restore operation. The system uses a client-server model in which data sources in a client store data to a cloud-based server that manages each data source as a separate namespace. The server process breaks up each data source into a series of fixed size sequential blocks, and uses a specialized change record type comprising reverse delta objects to synthesize full backup images from a first full image, wherein the full backup images are stored for every point in time in the object store. This provides a full backup to be available for client retrieval at any desired point in time. A simple block hash catalog is provided to reduce data duplication.

Abstract: Controlling data retention cost in cloud-based object storage by first obtaining current storage prices from the cloud provider for their different available storage tiers, and receiving user input for different possible retention parameters identity of data source and the retention period per storage tier. A cost calculator calculates the cost per time unit for each storage tier based on the current provider-specified price and the user-specified period. The calculated costs are displayed to the user a UI in tabular and/or graphical form. The user can change the retention period or data source parameters, and any impact of such changes on the cost will be calculated in real time and reflected in the graphical display. If the user changes the retention period to force an immediate movement of data from one tier to another, a data mover control facilitates this movement in the cloud object storage.

Abstract: A method for storing progress data while replicating a snapshot is disclosed. The operations comprise: determining blocks in a storage object that are different between a first and a second snapshots, wherein the first snapshot corresponds to a first tree structure, and the second snapshot corresponds to a second tree structure; and replicating the second snapshot to a remote storage system, comprising replicating all the blocks in the storage object determined to be different between the first and the second snapshots, wherein the replicating further comprises: upon completion of replication of each data slice, determining nodes in the first tree structure that reference only already-replicated blocks; orphaning all such nodes in the first tree structure; and subsequently, updating the first tree structure, comprising merging the first tree structure and the second tree structure, wherein the first tree structure is updated upon completion of replication of each data slice.

Abstract: Systems and methods for providing any point in time functionality. With a storage system such as a VSAN, any point in time protection is achieved by combining a metadata stream with snapshots of the storage system. This allows snapshots to be generated in hindsight such that any point in time functionality is provided.

Abstract: A method for storing progress data while replicating a snapshot is disclosed. The operations comprise: determining blocks in a storage object that are different between a first and a second snapshots, wherein the first snapshot corresponds to a first tree structure, and the second snapshot corresponds to a second tree structure; and replicating the second snapshot to a remote storage system, comprising replicating all the blocks in the storage object determined to be different between the first and the second snapshots, wherein the replicating further comprises: upon completion of replication of each data slice, determining nodes in the first tree structure that reference only already-replicated blocks; orphaning all such nodes in the first tree structure; and subsequently, updating the first tree structure, comprising merging the first tree structure and the second tree structure, wherein the first tree structure is updated upon completion of replication of each data slice.

Abstract: A computer implemented method, system, and computer program product comprising splitting IO directed to a LUN on a production site to a first thin LUN, determining to take a snapshot of the LUN at a point in time, reconfiguring the IO split from the LUN on the production site to be split to a second thin LUN, and copying changes from the first thin LUN to a protection storage device.

Abstract: In one aspect, a method includes intercepting write I/Os going to a volume on a storage array and sending the write I/Os to targets on a plurality of source-side data protection appliances based on a load balancing scheme. In another aspect, an article includes a non-transitory machine-readable medium that stores executable instructions. The instructions cause a machine to intercept write I/Os going to a volume on a storage array and send the write I/Os to targets on a plurality of source-side data protection appliances based on a load balancing scheme. In a further aspect, an apparatus includes circuitry configured to intercept write I/Os going to a volume on a storage array and send the write I/Os to targets on a plurality of source-side data protection appliances based on a load balancing scheme.

Abstract: Described embodiments provide systems and methods for synchronizing a production volume and a backup volume of a storage system. A first thin volume is created and associated with the production volume. A first replica of the production volume is generated by copying data from the production volume to a replica volume. During the copying, an I/O request to be written to the production volume may be received. Data from the I/O request is written to the first thin volume and data changed due to the I/O request is tracked in metadata associated with the production volume and the first thin volume. A size of the first thin volume is checked, and when the size of the first thin volume is below a threshold, changes from the first thin volume are applied asynchronously to the backup storage.

Abstract: In one aspect, a method includes separating a set of virtual machines from a first consistency group to a second consistency group and third consistency group. The method also includes combining a first virtual machine of the second consistency group to the third consistency group to form a fourth consistency group.

Abstract: A computer implemented method, a computer program product, and a system for replicating data comprising: intercepting the IO data to a volume at the splitter; sending the IO data intercepted at the splitter to the volume; sending metadata of the IO data to a production RPA from the splitter, wherein an acknowledgement is sent to the splitter indicating that the metadata was received by the production RPA; and sending IO data, metadata, and timestamp of IO data to a remote site from the splitter, wherein IO data is sent to the remote site after splitter receives acknowledge of production RPA receiving the IO data.

Abstract: In one embodiment, a method includes intercepting a write command to write to a volume at a production site, saving data from the write command to a cache, sending metadata from the write command to a data protection appliance (DPA), writing the data to the volume, receiving a request for the data in the cache, sending the data from cache to a data protection appliance (DPA) at the production site and sending the data from the DPA to the replication site.

Abstract: A computer program product, system, and method for generating coded fragments comprises receiving, at a primary data protection appliance (DPA), an I/O write for a user volume; determining a distributed consistency group (DCG) associated with the user volume, the DCG having a plurality of replica copies; determining one or more secondary DPAs assigned to one or more of the replica copies; sending the I/O write from the primary DPA to each of the secondary DPAs; and applying, at each of the secondary DPAs, the I/O write to at least one of the replica copies assigned to the secondary DPA.

Abstract: In one aspect, a method includes receiving, from a splitter, an I/O to be written from a virtual machine to a production virtual machine disk; sending an acknowledgement to the splitter that the I/O is received; writing data from the I/O to a first differential virtual machine disk; generating a snapshot; allocating a second differential virtual machine disk; redirecting new I/Os to the virtual machine, arriving from the splitter, to the second differential virtual machine disk and applying changes from first differential disk to a copy of data on a backup storage.

Abstract: In one aspect, a method includes making a first active buffer a passive buffer, generating a second active buffer to receive new I/Os, making a list of locations that need realignment in the passive buffer, flushing open I/Os at a splitter, reading I/Os which need realignment, discarding re-aligned I/Os for overwritten areas and sending the passive buffer to a replica site.

Abstract: In one aspect, a method includes receiving an I/O, incrementing a first counter in an active data structure in a backlog at a splitter after receiving the I/O, storing I/O metadata of the I/O in the active data structure, incrementing a second counter in the active data structure or a passive data structure in the backlog if the I/O was written to a storage array and received by the data protection appliance and incrementing a third counter in the active data structure or the passive data structure if either the I/O was not written to a storage array or not received by the data protection appliance.