Streaming and sequential data replication

Patent metadata

Field	Value
Publication number	US-10476955-B2
Application number	US-201715612241-A
Country	US
Kind code	B2
Filing date	Jun 2, 2017
Priority date	Jun 2, 2016
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

Systems and methods of capturing application data locally on a source data management system and performing a high performance, bandwidth optimized replication the data from the source data management system to a remote data management system either (a) in parallel with copying data from an application to the source data management system, or (b) serially by replicating the data to the remote data management system after the data is copied to the source data management system, wherein the determination to operate in parallel or serially is based on at least one of a characteristic of the data and user specified tuning parameters.

First claim

The invention claimed is:

1. A computerized method of capturing application data locally on a source data management system and performing a high performance, bandwidth optimized replication the data from the source data management system to a remote data management system either (a) in parallel with copying data from an application to the source data management system, or (b) serially by replicating the data to the remote data management system after the data is copied to the source data management system, wherein the determination to operate in parallel or serially is based on at least one of a characteristic of the data and user specified tuning parameters, the method comprising:
requesting, at a source computing device, application data from a host where an application is running, the application data comprising at least one of:
a full copy of the application data, and
changes associated with the application data, the changes including changed elements of the application data since a previous data capture of the application data from a host, and a list of changed elements associated with the application data;

receiving, by the source computing device, the application data;
determining, by the source computing device, a parameter associated with an ordering of the changed elements;
beginning to copy, by the source computing device, the application data to a location within the source computing device;
if the parameter is indicative of a non-overlapping order, beginning to replicate, by the source computing device, the application data to a remote computing device in parallel with copying the application data within the source computing device, the source computing device and the remote computing device being connected by a network; and
if the parameter is indicative of an overlapping order, beginning to replicate serially, by the source computing device, the application data to the remote computing device after copying the application data within the source computing device is complete,
thereby copying application data to a source data management system and replicating the application data from a source data management system to a remote data management system serially or in parallel based on at least one of a characteristic of the data and user specified tuning parameters.

2. The computerized method of claim 1 , wherein the full copy of the application data is requested when no previous captured version of the application data exists on the source computing device.

3. The computerized method of claim 2 , wherein the full copy is automatically associated with a parameter indicating a non-overlapping order, the computerized method further comprising replicating the full copy from the source computing device to the remote computing device in parallel with copying the full copy from the host to the source computing device.

4. The computerized method of claim 1 , wherein the list of changed elements is a list of changed extents within a block device, wherein each extent describes a range of modified bytes or modified blocks.

5. The computerized method of claim 4 , wherein replicating the application data from the source computing device to the remote computing device includes combining the extent list with each previous extent list associated with each previous data capture of the application data since a last successful replication operation.

6. The computerized method of claim 1 , wherein the list of changed elements is a bitmap, the bitmap including bit values of zero and one to represent whether a block or a grain within the application data is associated with a changed element, the grain being associated with a number of blocks.

7. The computerized method of claim 1 , further comprising:
storing, by the remote computing device, the application data alongside previously replicated application data on the remote computing device to create a complete representation of the changed application data and the previously replicated application data.

8. The computerized method of claim 1 , further comprising replicating, by the source computing device, additional application data to the remote computing device when the remote computing device includes no previously replicated application data from the source computing device.

9. The computerized method of claim 1 , wherein replicating the application data from the source computing device in parallel with copying the application data to the source computing device comprises:
a) copying, by the source computing system, an element of changed application data from the host application to the source computing device;
b) replicating, from the source computing device to the remote computing device, the element of changed application data; and
repeating steps a) and b) until each element of the changed application data is copied to the source computing device and replicated to the remote computing device.

10. The computerized method of claim 9 , wherein a speed associated with copying each of the elements to the source computing device is substantially similar to a speed associated with replicating to the remote computing device.

11. The computerized method of claim 9 , wherein a speed associated with replicating each of the elements to the remote data management system is substantially slower than a speed associated with copying each of the elements to the source computing device, the method further comprising:
determining, by the source computing device, all of the application data is copied to the source computing device;
stopping, by the source computing device, the replication of the application data to the remote staging disk in response to determining that all of the application data is copied to the source staging disk;
determining, by the source computing device, an offset associated with a data element of the application data being replicated prior to the stop; and
resuming, by the source computing device, replication of the application data occurring after the offset from the location associated with the copied application data being stored in the source computing device to the remote computing device.

12. The computerized method of claim 1 , further comprising:
determining, by the source computing device, a bandwidth level of the network, the bandwidth level indicative of the network being at least one of a high bandwidth network and a low bandwidth network;
enabling, by the source computing device, an application of deduplication and compression to the application data prior to replicating the application data to the remote computing device when the bandwidth level is indicative of the network being a low bandwidth network; and
disabling, by the source computing device, the application of deduplication and compression to the application data prior to replicating the application data to the remote computing device when the bandwidth level is indicative of the network being a high bandwidth network.

13. The computerized method of claim 1 , wherein a frequency associated with the copying of the application data to a location within the source computing device is higher than a frequency associated with replicating the application data from the source computing device to the remote computing device.

14. A computing system for capturing application data locally on a source data management system and performing a high performance, bandwidth optimized replication the data from the source data management system to a remote data management system either (a) in parallel with copying data from an application to the source data management system, or (b) serially by replicating the data

Assignees

• Actifio Inc

Inventors

• Mutalik Madhav
• Palaparthi Satya Sri Kanth
• GANTA Suresh

Classifications

• H04L43/0882

  Utilisation of link capacity · CPC title

• G06F11/2097

  maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups) · CPC title

• G06F11/2094

  Redundant storage or storage space (G06F11/2056 takes precedence) · CPC title

• G06F11/1464

  for networked environments · CPC title

• G06F11/1453

  using de-duplication of the data · CPC title