Active/active replication

Patent metadata

Field	Value
Publication number	US-9575857-B1
Application number	US-201213596180-A
Country	US
Kind code	B1
Filing date	Aug 28, 2012
Priority date	Jun 27, 2012
Publication date	Feb 21, 2017
Grant date	Feb 21, 2017

Abstract

A computer implemented method, system, and computer program for providing co-temporanious access to a single volume at a first site and a second site, the method comprising exposing the volume at the first site, exposing the volume at the second site; wherein the volume has the same identifier at the first site and the second site and wherein the write IO to the first volume is being replicated to the second volume, and redirecting IO to the second volume to the first volume.

First claim

What is claimed is:

1. A computer implemented method comprising:
providing contemporaneous access to a single volume at a first site and a second site, by:
exposing a first virtual volume corresponding to the single volume in a first hypervisor at the first site;
exposing a second virtual volume corresponding to the single volume in a second hypervisor at the second site; wherein the first virtual volume and the second virtual volume have the same identifier in the first hypervisor at the first site and in the second hypervisor at the second site and;
upon receiving a write IO to the first virtual volume from a virtual machine running in the first hypervisor:
replicating the write IO to the second volume using a first splitter on the first site; wherein the first splitter runs in a layer of the first hypervisor; wherein the replication of IOs from the first site to the second site includes the first splitter sending a copy of the write IO directed to the first virtual volume to the second site for writing to the second virtual volume; and

upon receiving an IO directed to the second virtual volume from a virtual machine running in the second hypervisor:
redirecting the IO sent to the second virtual volume to the first virtual volume; where the redirection of the IO is performed using a second splitter at the second site; wherein the second splitter runs in a layer of the second hypervisor; and,
determining if the IO is a write IO; and, based on a determination the IO is a write IO, sending a notification of the write IO at the first virtual volume to the virtual volume, and causing the second site applying the write IO based on the notification.

2. The method of claim 1 further comprising applying the write IO received from the second site to the volume at the first site.

3. The method of claim 1 wherein write IO directed to the second volume are kept in a cache at second site.

4. The method of claim 1 further comprising:
stopping access to the first volume at the first site; and
failing over to the second volume on the second site.

5. The method of claim 4 wherein failover comprises changing replication from the first site to the second site to synchronous and stopping redirection of IOs from second site to first site.

6. The method of claim 1 wherein the first splitter sends IO to be replicated to the second volume to a virtual Data Protection Appliance (vDPA) on the first site and wherein the vDPA on the first site send the IO to a second vDPA on the second site.

7. The method of claim 6 further comprising:
upon receipt of the second vDPA of an indication of an IO sent to the first site from the second site, storing data corresponding to the IO stored in a cache of the second vDPA in a journal at the second site in place of the indication received from the first site.

8. The method of claim 7 wherein data from the journal representing writes to the single volume is applied to the second volume.

9. A computer program product for providing contemporaneous access to a single volume at a first site and a second site comprising:
a non-transitory computer readable medium encoded with computer executable program code for replication of data, the computer executable program code executed by one or more processor for providing contemporaneous access to a single volume at a first site and a second site, by performing:
exposing a first virtual volume corresponding to the single volume in a first hypervisor at the first site;
exposing a second virtual volume corresponding to the single volume in a second hypervisor at the second site; wherein the first virtual volume and the second virtual volume have the same identifier in the first hypervisor at the first site and in the second hypervisor at the second site and;
upon receiving a write IO to the first virtual volume from a virtual machine running in the first hypervisor:
replicating the write IO to the second volume using a first splitter on the first site; wherein the first splitter runs in a layer of the first hypervisor; wherein the replication of IOs from the first site to the second site includes the first splitter sending a copy of the write IO directed to the first virtual volume to the second site for writing to the second virtual volume; and

upon receiving an IO directed to the second virtual volume from a virtual machine running in the second hypervisor:
redirecting the IO sent to the second virtual volume to the first virtual volume; where the redirection of the IO is performed using a second splitter at the second site; wherein the second splitter runs in a layer of the second hypervisor; and,
determining if the IO is a write IO; and, based on a determination the IO is a write IO, sending a notification of the write IO at the first virtual volume to the virtual volume, and causing the second site applying the write IO based on the notification.

10. The computer program product of claim 9 wherein the computer executable program code is further performing:
applying the write IO received from the second site to the volume at the first site.

11. The computer program product of claim 9 wherein write IO directed to the second volume are kept in a cache at second site.

12. The computer program product of claim 9 wherein the computer executable program code is further performing:
stopping access to the second volume at the first site; and
failing over to the second volume on the second site.

13. The computer program product of claim 12 wherein the computer executable program code is further performing:
moving replication from first site to second site to be synchronous; and
stopping redirection of IOs from second site to first site.

14. A system for providing contemporaneous access to a single volume at a first site and a second site, the system comprising:
one or more processors;
one or more memories;
a first hypervisor at the first site; wherein the first hypervisor is enabled to run one or more virtual machines; wherein the first hypervisor has a first splitter;
a second hypervisor at the second site; wherein the second hypervisor is enabled to run one or more virtual machines; wherein the second hypervisor has a second splitter; and
computer-executable program loaded in the one or more memories, wherein the computer-executable program is executed by the one or more processors to perform providing contemporaneous access to a single volume at a first site and a second site, by:
exposing a first virtual volume corresponding to the single volume in the first hypervisor at the first site;
exposing a second virtual volume corresponding to the single volume in the second hypervisor at the second site; wherein the first virtual volume and the second virtual volume have the same identifier in the first hypervisor at the first site and in the second hypervisor at the second site and;
upon receiving a write IO to the first virtual volume from a virtual machine running in the first hypervisor:
replicating the write IO to the second volume using a first splitter on the first site; wherein the first splitter runs in a layer of the first hypervisor; wherein the replication of IOs from the first site to the second site includes the first splitter sending a copy of the write IO directed to the first virtual volume to the second site for writing to the second virtual volume; and

upon receiving an IO directed to the second virtual volume from a virtual machine running in the second hypervisor:
redirecting the IO sent to the second virtual volume to the first virtual volume; where the redirection of the IO is performed using a second splitter at the

Assignees

• Natanzon Assaf
• Emc Ip Holding Co Llc

Inventors

• Natanzon Assaf

Classifications

• G06F11/2082

  Data synchronisation · CPC title

• G06F11/2064 Primary

  while ensuring consistency · CPC title

• G06F2201/855

  Details of asynchronous mirroring using a journal to transfer not-yet-mirrored changes · CPC title

• G06F16/178

  Techniques for file synchronisation in file systems · CPC title

• G06F2201/815

  Virtual · CPC title