Systems and methods for continuous data protection comprising storage of completed I/O requests intercepted from an I/O stream using touch points
Example embodiments relate generally to systems and methods for continuous data protection (CDP) and more specifically to an input and output (I/O) filtering framework and log management system to seek a near-zero recovery point objective (RPO).
1. A method for continuous data protection for a virtual machine (VM) having a virtual disk, comprising:
accessing, at an interception point in an input/output I/O path, an I/O stream comprising records of completed I/O requests and records of uncompleted I/O requests exchanged between the VM and a virtualization server;
configuring an I/O touch point in the I/O stream, the I/O touch point associated with I/O request cancellation of uncompleted I/O requests;
filtering the I/O stream to intercept only the records of the completed I/O requests in the I/O stream using the I/O touch point;
storing, from among the records of the completed I/O requests and the records of the uncompleted I/O requests, only the records of the completed I/O requests at a backup site based at least in part on the filtering;
forming a recoverable snapshot-log chain by associating the records stored at the backup site with a snapshot;
receiving a request for recoverable data from a replication target; and
sending data to the replication target based at least on a portion of the recoverable snapshot-log chain.
2. The method of claim 1 , further comprising:
establishing a filter framework at the interception point, the filter framework including an I/O stack and an I/O filter.
3. The method of claim 2 , wherein the virtualization server is an ESX hypervisor server, the method further comprising:
including a filter driver for the filter framework within the ESX hypervisor server.
4. The method of claim 2 , further comprising:
configuring a filter driver to intercept only the completed I/O requests in the I/O stream using the I/O touch point, wherein the snapshot is a full image snapshot.
5. The method of claim 4 , wherein the I/O request cancellation is a distributed I/O request cancellation of a plurality of I/O requests comprising a first set of completed I/O requests and a second set of uncompleted I/O requests, the method further comprising:
managing the distributed I/O request cancellation at the I/O stack and not at the I/O filter of the filter framework.
6. A system for continuous data protection for a virtual machine (VM) having a virtual disk, comprising:
at least one processor for executing machine-readable instructions; and
a memory storing instructions executable by the at least one processor to cause the system to:
access, at an interception point in an input/output I/O path, an I/O stream comprising records of completed I/O requests and records of uncompleted I/O requests exchanged between the VM and a virtualization server;
configure an I/O touch point in the I/O stream, the I/O touch point associated with I/O request cancellation of uncompleted I/O requests;
filter the I/O stream to intercept only the records of the completed I/O requests in the I/O stream using the I/O touch point;
store, from among the records of the completed I/O requests and the records of the uncompleted I/O requests, only the records of the completed I/O requests at a backup site based at least in part on the filtering;
form a recoverable snapshot-log chain by associating the records stored at the backup site with a snapshot;
receive a request for recoverable data from a replication target; and
send data to the replication target based at least on a portion of the recoverable snapshot-log chain.
7. The system of claim 6 , wherein the instructions are further executable by the at least one processor to cause the system to:
establish a filter framework at the interception point, the filter framework including an I/O stack and an I/O filter.
8. The system of claim 7 , wherein the virtualization server is an ESX hypervisor server, and wherein the instructions are further executable by the at least one processor to cause the system to:
include a filter driver for the filter framework within the ESX hypervisor server.
9. The system of claim 7 , wherein the instructions are further executable by the at least one processor to cause the system to:
configure a filter driver to intercept only the completed I/O requests in the I/O stream using the I/O touch point, wherein the snapshot is a full image snapshot.
10. The system of claim 9 , wherein the I/O request cancellation is a distributed I/O request cancellation of a plurality of I/O requests comprising a first set of completed I/O requests and a second set of uncompleted I/O requests, and wherein the instructions are further executable by the at least one processor to cause the system to:
manage the distributed I/O request cancellation at the I/O stack and not at the I/O filter of the filter framework.
11. A non-transitory, machine-readable medium storing instructions for continuous data protection for a virtual machine (VM) having a virtual disk, wherein the instructions, when read by a machine, cause the machine to:
access, at an interception point in an input/output I/O path, an I/O stream comprising records of completed I/O requests and records of uncompleted I/O requests exchanged between the VM and a virtualization server;
configure an I/O touch point in the I/O stream, the I/O touch point associated with I/O request cancellation of uncompleted I/O requests;
filter the I/O stream to intercept only the records of the completed I/O requests in the I/O stream using the I/O touch point;
store, from among the records of the completed I/O requests and the records of the uncompleted I/O requests, only the records of the completed I/O requests at a backup site based at least in part on the filtering;
form a recoverable snapshot-log chain by associating the records stored at the backup site with a snapshot;
receive a request for recoverable data from a replication target; and
send data to the replication target based at least on a portion of the recoverable snapshot-log chain.
12. The non-transitory, machine-readable medium of claim 11 , wherein the instructions, when read by the machine, further cause the machine to:
establish a filter framework at the interception point, the filter framework including an I/O stack and an I/O filter.
13. The non-transitory, machine-readable medium of claim 12 , wherein the virtualization server is an ESX hypervisor server, and wherein the instructions, when read by the machine, further cause the machine to:
include a filter driver for the filter framework within the ESX hypervisor server.
14. The non-transitory, machine-readable medium of claim 12 , wherein the instructions, when read by the machine, further cause the machine to:
configure a filter driver to intercept only the completed I/O requests in the I/O stream using the I/O touch point, and wherein the snapshot is a full image snapshot.
15. The non-transitory, machine-readable medium of claim 14 , wherein the I/O request cancellation is a distributed I/O request cancellation of a plurality of I/O requests comprising a first set of completed I/O requests and a second set of uncompleted I/O requests, and wherein the instructions, when read by the machine, further cause the machine to:
manage the distributed I/O request cancellation at the I/O stack and not at the I/O filter of the filter framework.