Redundancy Method, Device, and System

Patent metadata

Field	Value
Publication number	US-2018217903-A1
Application number	US-201815938609-A
Country	US
Kind code	A1
Filing date	Mar 28, 2018
Priority date	Sep 29, 2015
Publication date	Aug 2, 2018
Grant date	—

Abstract

A redundancy method includes that a first disaster management function (DMF) device on a first site side receives a first request including identification information of a first virtual machine (VM) and a recovery point objective (RPO), allocates a maximum allowable delay time to each node that input/output (TO) data of the first VM passes through in a redundancy process, and sends a second request to a second DMF device on a second site side. The second request includes a maximum allowable delay time of a second replication gateway function (RGF) device on the second site side, and a maximum allowable delay time of an IO writer function (IOWF) device on the second site side and requests the second site side to perform redundancy on the first VM. Hence, the RPO requirements of the tenants can be satisfied in an entire redundancy process.

First claim

What is claimed is:

1 . A redundancy method, comprising:
receiving, by a first disaster management function (DMF) device on a first site side, a first request from a cloud business function (CBF) device, wherein the first request requests to perform redundancy on a first virtual machine (VM) on the first site side, and wherein the first request comprises identification information of the first VM and a recovery point objective (RPO); allocating, by the first DMF device according to the RPO, a maximum allowable delay time to each node that input/output (TO) data of the first VM passes through in a redundancy process; and sending, by the first DMF device, a second request to a second DMF device on a second site side, wherein the second request requests the second site side to perform the redundancy on the first VM, and wherein the second request comprises the identification information of the first VM, a maximum allowable delay time of a second replication gateway function (RGF) device on the second site side, and a maximum allowable delay time of an IO writer function (IOWF) device on the second site side.

2 . The method according to claim 1 , wherein allocating the maximum allowable delay time to each node comprises:
determining, by the first DMF device, a maximum allowable transmission delay per unit bandwidth on a link for transmitting the IO data of the first VM; determining, by the first DMF device, a difference between the RPO and the maximum allowable transmission delay per unit bandwidth on the link for transmitting the IO data of the first VM; setting, by the first DMF device, partial duration of the difference as a reserved delay; and allocating the maximum allowable delay time to each node based on remaining duration of the difference, wherein a sum of the maximum allowable delay times of nodes comprises the remaining duration of the difference.

3 . The method according to claim 2 , wherein a maximum allowable delay time allocated by the first DMF device to a node, that is configured to remotely transmit IO data, of the nodes is greater than a maximum allowable delay time allocated by the first DMF device to a node, that is configured to locally transmit IO data, of the nodes, and wherein the maximum allowable delay time allocated by the first DMF device to the node, that is configured to locally transmit the IO data, of the nodes is greater than a maximum allowable delay time allocated by the first DMF device to a node, that is configured to cache IO data, of the nodes.

4 . The method according to claim 1 , further comprising sending, by the first DMF device, a third request to a first virtualization management function (VMF) device on the first site side, and wherein the third request requests the first VMF device to configure an IO catcher function (IOCF) device on the first site side.

5 . The method according to claim 4 , further comprising instructing, by the first DMF device after receiving a timeout request from the IOCF device, the IOCF device to discard IO data that needs to be transmitted currently.

6 . The method according to claim 4 , further comprising:
reallocating, by the first DMF device, a maximum allowable delay time to the IOCF device after receiving a timeout request from the IOCF device; and sending the reallocated maximum allowable delay time to the IOCF device.

7 . The method according to claim 1 , further comprising sending, by the first DMF device, a fourth request to a first RGF device on the first site side, wherein the fourth request requests the first RGF device to receive the IO data of the first VM intercepted by an IO catcher function (IOCF) device and send the received IO data to the second RGF device on the second site side according to a maximum allowable delay time allocated by the first DMF device to the first RGF device.

8 . The method according to claim 7 , further comprising:
reallocating, by the first DMF device, the maximum allowable delay time to the first RGF device after receiving a timeout request from the first RGF device; and sending the reallocated maximum allowable delay time to the first RGF device.

9 . The method according to claim 1 , further comprising:
reallocating, by the first DMF device after receiving a timeout request from the second DMF device on the second site side, the maximum allowable delay time to the second RGF device on the second site side and the maximum allowable delay time to the IOWF device on the second site side according to the timeout request; and sending the reallocated maximum allowable delay time to the second DMF device.

10 . The method according to claim 1 , further comprising:
reallocating, by the first DMF device after receiving a timeout request from the second DMF device on the second site side, the maximum allowable delay time to the second RGF device on the second site side according to the timeout request; and sending the reallocated maximum allowable delay time to the second DMF device.

11 . The method according to claim 1 , further comprising:
reallocating, by the first DMF device after receiving a timeout request from the second DMF device on the second site side, the maximum allowable delay time to the IOWF device on the second site side according to the timeout request; and sending the reallocated maximum allowable delay time to the second DMF device.

12 . The method according to claim 1 , further comprising:
reallocating, by the first DMF device, the maximum allowable delay time to each node that the IO data of the first VM passes through in the redundancy process after a transmission rate on a link for transmitting the IO data of the first VM changes; and notifying nodes.

13 . A redundancy method, comprising:
receiving, by a second disaster management function (DMF) device on a second site side, a second request from a first DMF device on a first site side, wherein the second request requests the second site side to perform redundancy on a first virtual machine (VM) on the first site side; sending, by the second DMF device, a seventh request to a second virtualization management function (VMF) device on the second site side, wherein the seventh request requests the second VMF device to configure an input/output (TO) writer function (IOWF) device on the second site side; and sending, by the second DMF device, an eighth request to a second replication gateway function (RGF) device on the second site side, wherein the eighth request requests the second RGF device to locally transmit IO data of the first VM according to a maximum allowable delay time allocated by the first DMF device to the second RGF device.

14 . The method according to claim 13 , further comprising:
instructing, by the second DMF device after receiving a timeout request from the IOWF device, the IOWF device to discard IO data that needs to be transmitted currently; or sending, by the second DMF device after receiving the timeout request from the IOWF device, the timeout request to the first DMF device to request the first DMF device to reallocate a maximum allowable delay time to the IOWF device, and sending the reallocated maximum allowable delay time, from the first DMF device, to the IOWF device after receiving the reallocated maximum allowable delay time.

15 . The method according to claim 14 , further comprising:
sending, by the second DMF device after receiving a timeout request from the second RGF device, the timeout request to the first DMF device to request the first DMF device to reallocate the maximum allowable delay time to the second RGF device; and sending the reallocated ma

Assignees

• Huawei Tech Co Ltd

Inventors

• Wu Xiangyang
• Zhao Zhi
• Chen Rong
• Jiang Xuan

Classifications

• G06F2009/45579

  I/O management, e.g. providing access to device drivers or storage · CPC title

• H04L41/0663

  Performing the actions predefined by failover planning, e.g. switching to standby network elements · CPC title

• H04L43/0852

  Delays · CPC title

• H04L41/5051 Primary

  Service on demand, e.g. definition and deployment of services in real time · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title