System and method to proactively maintain a consistent recovery point objective (RPO) across data centers

• US 9,489,268 B2
• Filed: 01/27/2014
• Issued: 11/08/2016
• Est. Priority Date: 04/01/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: Sanovi Technologies PVT Limited (Christian Nicholas Rosenkrantz Stadil,)
• Inventors: Babashetty, Sharanabasappa, Garg, Devendra, Vonna, Rajasekhar, Madawat, Sukumar
• Primary Examiner(s): Pham, Tuan A
• Application Number:
  US14/164,665
• Publication Number:
  US 20140297588A1
• Time in Patent Office: 1,016 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 11/302 where the computing system component is a software system
• CPC Class Codes: G06F 11/3027 where the computing system component is a bus
• CPC Class Codes: G06F 2201/81 Threshold

Abstract

A system and method for proactively monitoring and maintaining a consistent recovery point objective (RPO) across data centers, the system comprising: one or more RPO Management Server(s) logically connected to one or more Production Sites and one or more Disaster Recovery Sites; a Network connecting the said RPO Management Server(s) with the said Production Site and the said Disaster Recovery Site wherein the said RPO Management Server is provided with at least one RPO Manager, at least one Disaster Recovery Management (DRM) System and at least one Replication Management System.

Claims

1. A method for proactively monitoring and maintaining a consistent recovery point objective (RPO) across data centers comprising a plurality of computers, the method comprising the steps of:

providing a plurality of different RPO computation methods including a data RPO computation for monitoring and/or computing RPO of only critical components of applications and an application RPO computation for monitoring and/or computing RPO from both critical and non-critical components of applications for use by a RPO manager executing on an RPO management server comprising one or more computers;

RPO Manager executing on the RPO management server selecting an appropriate RPO computation method for the system from the plurality of different RPO computation methods;

the RPO Manager executing on the RPO management server fetching the input RPO or threshold RPO for the application(s);

the RPO manager executing on the RPO management server requesting the input RPO or threshold RPO from the Disaster Recovery Management (DRM) System via a network interface;

the RPO manager executing on the RPO management server requesting the current RPO for the application(s) from the Disaster Recovery Management (DRM) System;

the Disaster Recovery Management System sending the current RPO for the application(s) to the RPO manager executing on the RPO management server;

the RPO Manager executing on the RPO management server determining if the current RPO of the application(s) is closer to the configured threshold RPO;

the RPO manager executing on the RPO management server checking whether the RPO computation method allocated during scheduled window of EOD processing or administration operations is application RPO computation;

the RPO Manager executing on the RPO management server pausing the replication of non-critical components of application(s) if the current RPO of the application(s) is closer to the said configured threshold RPO and the said allocated RPO computation method is application RPO computation; and

the RPO Manager executing on the RPO management server resuming the replication of critical components of application(s) if the current RPO of the application(s) is closer to the said configured threshold RPO and the said allocated RPO computation method is application RPO computation.

2. The method as claimed in claim 1, the method further comprising the steps of:

(a) prompting the RPO Manager to fetch the input RPO or threshold RPO and current RPO for the application(s) from the Disaster Recovery Management (DRM) System;

(b) prompting the RPO Manager to fetch the priority of application(s) from Disaster Recovery Management (DRM) System;

(c) prompting the RPO Manager to fetch replicator data transfer rate for the application(s) from the Replication Management System;

(d) prompting the RPO Manager to fetch replicator data lag reading from the said Replication Management System;

(e) prompting the RPO Manager to calculate the RPO deviation for the said application(s);

(f) prompting the RPO Manager to check application(s) having current RPO greater than or equal to “

k”

percentage of input RPO (where “

k”

is a particular value);

(g) prompting the RPO Manager to shortlist top “

n”

applications checked at step (f);

(h) prompting the RPO Manager to check for low priority application(s) from the a number “

n”

of shortlisted application(s) with paused replication rates and huge data lag;

(i) prompting the RPO Manager to resume the replication of low priority applications checked at step (h);

(j) prompting the RPO Manager to resume the process from the step of selection of an appropriate computation method for the system or from the step of fetching the current RPO and threshold RPO for applications by RPO Manager subsequent to the replication of low priority applications at step (i).

3. The method as claimed in claim 2, the method further comprising the steps of:

(a) prompting the RPO Manager to check application with high data transfer rate and current RPO value below “

M”

percentage of the said RPO threshold value (where “

M”

is a particular value) if no low priority applications are obtained from shortlisted “

n”

applications;

(b) prompting the RPO Manager to resume the management of RPO levels from selection of an appropriate window for End of Day (EOD) processing if no applications found at step (a) with high data transfer rate and current RPO value below “

M”

percentage of the threshold value;

(c) prompting the RPO manager to shortlist bottom “

N”

applications (where “

N”

is a number of such applications) if applications found at step (a) have high data transfer rate and current RPO value below “

M”

percentage of the said RPO threshold value;

(d) prompting the RPO Manager to check the shortlisted “

N”

applications for high priority applications;

(e) prompting the RPO Manager to check for application(s) peak hours;

(f) prompting the RPO Manager to alter the network bandwidth utilization parameters of replicators for high priority applications checked during non-peak application(s) hours;

(g) prompting the RPO Manager to resume the process from the step of selection of an appropriate computation method for the system or from the step of fetching current RPO and threshold RPO for applications for applications by the RPO Manager after the said network bandwidth utilization parameters of replicators are altered at step (f);

(h) prompting the RPO Manager to not alter the network bandwidth utilization parameters of replicators for high priority applications during peak application(s) hours;

(i) prompting the RPO Manager to resume the process from the step of selection of an appropriate computation method for the system or from the step of fetching current RPO and threshold RPO for applications by the RPO Manager after the said network bandwidth utilization parameters of replicators are not altered at step (h).

4. The method as claimed in claim 3, the method further comprising the steps of:

(a) prompting the RPO Manager to check the shortlisted “

N”

application(s) for low priority applications if no high priority application(s) are found;

(b) prompting the RPO manager to calculate application write rate for low priority applications found at step (a);

(c) prompting the RPO Manager to initiate data compression if the application write rate for said low priority application(s) is greater than or equal to allocated network bandwidth for the said low priority application(s);

(d) prompting the RPO Manager to resume the process from the step of selection of an appropriate computation method for the system or from the step of fetching current RPO and threshold RPO for applications by the RPO Manager after the said compression of data at step (c);

(e) prompting the RPO Manager to pause the replication of the said low priority application(s) if the application write rate is lesser than or equal to allocated network bandwidth for the said low priority application(s);

(f) prompting the RPO Manager to resume the process from the step of selection of an appropriate computation method for the system or from the step of fetching current RPO and threshold RPO for applications by the RPO Manager after the said replication of the said low priority application(s) at step (e) is paused.

5. The method as claimed in claim 2, wherein the value of “

k”

, “

n”

, “

N” and

“

M”

are configurable.

6. The method as claimed in claim 5, wherein the value of “

n” and

“

N”

is equal to 3.

7. The method as claimed in claim 5, wherein the value of “

k”

percent is 90%.

8. The method as claimed in claim 5, wherein the value of “

M”

percent is 50%.

9. A method for proactively monitoring and maintaining a consistent recovery point objective (RPO) across data centers comprising a plurality of computers, the method comprising the steps of:

providing a plurality of different RPO computation methods including a data RPO computation for monitoring and/or computing RPO of only critical components of applications and an application RPO computation for monitoring and/or computing RPO from both critical and non-critical components of applications for use by a RPO manager executing on an RPO management server comprising one or more computers;

RPO Manager executing on the RPO management server selecting an appropriate RPO computation method for the system from the plurality of different RPO computation methods;

the RPO Manager executing on the RPO management server fetching the input RPO or threshold RPO for the application(s);

the RPO manager executing on the RPO management server requesting the input RPO or threshold RPO from Disaster Recovery Management (DRM) System comprising a plurality of computers interacting with the RPO manager via a network interface;

the RPO manager executing on the RPO management server requesting a current RPO for the application(s) from the Disaster Recovery Management (DRM) System;

the Disaster Recovery Management System sending the current RPO for the application(s) to the RPO manager executing on the RPO management server;

the RPO Manager executing on the RPO management server determining if the current RPO of the application(s) is closer to the configured threshold RPO;

the RPO manager executing on the RPO management server checking whether the RPO computation method allocated during scheduled window of EOD processing or administration operations is application RPO computation;

the RPO Manager executing on the RPO management server pausing the replication of non-critical components of application(s) if the current RPO of the application(s) is closer to the said configured threshold RPO and the said allocated RPO computation method is application RPO computation; and

the RPO Manager executing on the RPO management server resuming the replication of critical components of application(s) if the current RPO of the application(s) is closer to the said configured threshold RPO and the allocated RPO computation method is application RPO computation, wherein

the RPO Manager executing on the RPO management server provides scheduler based pause/resume capabilities for RPO computation for end of Day (EOD) processing and for application administration window.

10. A method for proactively monitoring and maintaining a consistent recovery point objective (RPO) across data centers comprising a plurality of computers, the method comprising the steps of:

providing a plurality of different RPO computation methods including a data RPO computation for monitoring and/or computing RPO of only critical components of applications and an application RPO computation for monitoring and/or computing RPO from both critical and non-critical components of applications for use by a RPO manager executing on an RPO management server comprising one or more computers system;

RPO Manager executing on the RPO management server selecting an appropriate RPO computation method for the system from the plurality of different RPO computation methods;

prompting the RPO Manager executing on the RPO management server fetching the input RPO or threshold RPO for the application(s);

the RPO manager executing on the RPO management server requesting the input RPO or threshold RPO from the Disaster Recovery Management (DRM) System comprising a plurality of computers interacting with the RPO manager via a network interface;

the RPO manager executing on the RPO management server requesting the current RPO for the application(s) from the Disaster Recovery Management (DRM) System;

the Disaster Recovery Management System sending the current RPO for the application(s) to the RPO manager executing on the RPO management server;

the RPO Manager executing on the RPO management server determining if the current RPO of the application(s) is closer to the configured threshold RPO;

the RPO manager executing on the RPO management server checking whether the RPO computation method allocated during scheduled window of EOD processing or administration operations is application RPO computation;

the RPO Manager executing on the RPO management server pausing the replication of non-critical components of application(s) if the current RPO of the application(s) is closer to the configured threshold RPO and the allocated RPO computation method is application RPO computation; and

the RPO Manager executing on the RPO management server resuming the replication of critical components of application(s) if the current RPO of the application(s) is closer to the said configured threshold RPO and the said allocated RPO computation method is application RPO computation, wherein

the RPO Manager executing on the RPO management server is configured to provide intelligent RPO computation methods for application(s).