Data synchronization

A computer program product comprising:

• a non-transitory computer readable medium encoded with computer executable program code, wherein the code enables execution across one or more processors of:
• reading at least one signature of data of a production virtual machine on production site;
• wherein reading the at least one signature includes mapping a virtual machine disk offset into a physical storage offset and reading the signature from the physical storage;
• reading at least one signature of data of a replica virtual machine used to replicate the production virtual machine on a replica site wherein reading the at least one signature includes mapping a virtual machine disk offset into a physical storage offset and reading the signature from the physical storage;
• based on determination the signatures are not identical marking the locations for synchronization in a synchronization structure;
• wherein a snapshot of a Logical Unit (LU) containing a virtual machine file system containing the virtual machine disks of the production virtual machine and replica virtual machine is created on the production site and replica site;
• wherein the LU snapshot is parsed to find a list of Virtual Machine Disk (VMDK) blocks and a mapping of the VMDK blocks to LU blocks, wherein the VMDK blocks provided on the replication appliance comprise a thick VMDK so that mapping from the VMDK blocks to LU addresses exist;
• wherein a splitter intercepts Inputs/Outputs (IOs) for the virtual machine disk at the production site, and marks locations arriving to the virtual machine disk as dirty in a synchronization structure;
• and
• performing a failover to change a direction of replicate data flow from the replica site to the production site.

A system comprising:

• a production site;
• a replica site;
• and
• computer-executable logic operating in memory, wherein the computer-executable program logic is configured to enable execution across one or more processors of:
• reading signatures of data of the production virtual machine on production site;
• wherein reading the signature includes mapping a virtual machine disk offset into a physical storage offset and reading the signature from the physical storage;
• reading signatures of data of a replica virtual machine used to replicate the production virtual machine on the replica site wherein reading the signature includes mapping a virtual machine disk offset into a physical storage offset and reading the signature from the physical storage;
• based on determination the signatures are not identical marking the locations for synchronization in a synchronization structure;
• wherein a snapshot of a Logical Unit (LU) containing a virtual machine file system containing the virtual machine disks of the production virtual machine and replica virtual machine is created on the production site and replica site;
• wherein the LU snapshot is parsed to find a list of Virtual Machine disk (VMDK) blocks and a mapping of the VMDK blocks to LU blocks, wherein the VMDK blocks provided on the replication appliance comprise a thick VMDK so that mapping from the VMDK blocks to LU addresses exist;
• wherein a splitter intercepts Inputs/Outputs (IOs) for the virtual machine disk at the production site, and marks locations arriving to the virtual machine disk as dirty in a synchronization structure;
• and
• performing a failover to change a direction of replicate data flow from the replica site to the production site.