Outage window scheduler tool

• US 9,372,734 B2
• Filed: 08/27/2013
• Issued: 06/21/2016
• Est. Priority Date: 08/27/2013
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Andreatta, Richard Gerard, Madigan, David Joseph
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US14/011,472
• Publication Number:
  US 20150066813A1
• Time in Patent Office: 1,029 Days
• Field of Search: 706/11
• US Class Current: 1/1
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 9/505 considering the load

Abstract

An apparatus for determining a future outage window in which to perform work on a server may include an input for receiving historical performance data about at least one server, a non-transitory memory and a processor communicatively coupled to the input and the memory. The processor may be configured to use instructions stored in the memory to predict one or more likely future time windows to be recommended as possible outage windows in which to perform work on the server. The processor may analyze historical performance data about the at least one server to determine one or more historical time windows corresponding to low CPU utilization of the at least one server. In some cases, the processor may then predict one or more future time windows to be recommended for use as the outage window using, at least in part, the determined historical time windows.

Claims

1. An apparatus, comprising:

an input configured to receive information about at least one server;

a non-transitory memory storing computer-executable instructions;

a processor communicatively coupled to the input and the memory, the processor configured to execute the computer-executable instructions, wherein execution of the instructions would cause the processor to;

analyze the information associated with the at least one server to identify one or more historical time windows corresponding to low utilization of the at least one server, the historical time windows corresponding to low utilization being identified by comparing values of a plurality of processor utilization parameters to baseline values of the processor utilization parameters, the plurality of processor utilization parameters including at least;

a CPU user utilization including a percentage of users utilizing one or more applications on a particular server, and a CPU utilization, the baseline values being based, at least in part, on an age of the at least one server; and

predict, using at least the historical time windows, one or more future time windows corresponding to low utilization of the at least one server;

recommend, via a user interface, one or more of the future time windows for use as an outage window in which to perform work on the one or more servers, wherein each of the one or more future time windows is associated with a risk tier that corresponds to a risk of scheduling an outage of the one or more servers during each of the one or more future time windows, and the recommending includes filtering the one or more future time windows to display, via the user interface, only future time windows having a risk tier meeting a predetermined threshold; and

receive, via a user input of the user interface screen, a selection of one or more of the future time windows for use as the future outage window.

2. The apparatus of claim 1, wherein execution of the instructions would further cause the processor to determine the risk tier associated with each of the one or more future time windows.

3. The apparatus of claim 2, wherein the risk tier corresponds to a utilization level of the same server during one or more similar historical time windows.

4. The apparatus of claim 3, wherein the risk tier corresponds to whether the utilization level of the at least one server during one or more similar historical time windows is less than or equal to a specified threshold and/or a count of similar historical time windows having a utilization level of the at least one server less than or equal to the specified threshold.

5. The apparatus of claim 1, wherein the user interface is further configured to receive information about the at least one server and/or for presenting information about the one or more future time windows.

6. The apparatus of claim 5, wherein the user interface is displayed on a device separate from the apparatus and the device is configured to communicate with the apparatus via the Internet.

7. The apparatus of claim 5, wherein the user interface is further configured to facilitate entry of one or more conditions used when predicting the one or more future time window, wherein the one or more conditions include at least one of a time of day, a window duration, a day of the week, a maximum number of windows per day, and a maximum allowed confidence level.

8. The apparatus of claim 1, wherein the information received at the input includes historical operational information about one or more servers on a network and/or other server characteristics including an age of the server, a processor and one or more characteristics of the processor and/or memory.

9. The apparatus of claim 8, wherein the historical operational information includes historical processor utilization information for a specified time period.

10. The apparatus of claim 8, wherein the historical operational information is collected by a performance monitoring tool.

11. The apparatus of claim 8, wherein the user interface is further configured for specifying a number of days of historical information to be analyzed when predicting the one or more future time windows.

12. The apparatus of claim 11, wherein the number of days of historical information to be used is between 14 days of historical information to about 120 days of historical information.

13. A system comprising:

a first memory unit storing a database including historical processor utilization information about the operation of one or more servers on a network;

a scheduler coupled to the database, the scheduler including a second memory unit storing instructions and a processor, wherein execution of the instructions would cause the processor to;

analyze the processor utilization information associated with the one or more servers received from the database to identify one or more historical time windows corresponding to low utilization of the one or more servers, the historical time windows corresponding to low utilization being identified by comparing values of a plurality of processor utilization parameters to baseline values of the processor utilization parameters, the plurality of processor utilization parameters including at least;

a CPU user utilization including a percentage of users utilizing one or more applications on a particular server, and a CPU utilization, the baseline values being based, at least in part, on an age of the at least one server;

predict, using at least the historical time windows, one or more future time windows corresponding to low utilization of the one or more servers;

determine a ranking associated with each of the one or more future time windows, wherein each ranking corresponds to a risk tier associated with scheduling an outage of the one or more servers during the associated future time window;

recommend, via a user interface and using the rankings, one or more of the future time windows for use as an outage window in which to perform work on the one or more servers, wherein each of the one or more future time windows is associated with a risk tier that corresponds to a risk of scheduling an outage of the one or more servers during each of the one or more future time windows, and the recommendation includes filtering the one or more future time windows to display, via the user interface, only future time windows having a ranking meeting a predetermined threshold; and

receive, via a user input of the user interface screen, a selection of one or more of the future time windows for use as the future outage window.

14. The system of claim 13, further comprising the one or more servers on the network.

15. The system of claim 13, further comprising a performance analyzer for collecting historical performance data about the one or more servers, the performance data including the historical utilization information about the one or more servers.

16. A method comprising:

receiving, from a data store via a network, historical performance data about the utilization of a server;

analyzing, by a scheduler, the historical performance data about the utilization of the server to identify one or more historical time windows corresponding to low utilization of the server, the historical time windows corresponding to low utilization being identified by comparing values of a plurality of processor utilization parameters to baseline values of the processor utilization parameters, the plurality of processor utilization parameters including at least;

a CPU user utilization including a percentage of users utilizing one or more applications on the server at a particular time, and a CPU utilization at a particular time, the baseline values being based, at least in part, on an age of the server;

predicting, by the scheduler, one or more future time windows of low utilization using the historical performance data;

determining, by the scheduler, a ranking associated with each of the one or more future time windows, wherein each ranking corresponds to a risk tier associated with scheduling an outage of the one or more servers during the associated future time window

recommending, via a user interface, at least one of the future time windows to be used as a future outage window of the server, wherein the recommending includes filtering the one or more future time windows to display only future time windows having a risk tier under a defined threshold; and

receiving, via a user input of the user interface screen, a selection of one or more of the future time windows for use as the future outage window.

17. The method of claim 16, wherein the one or more historical time windows are used for predicting the one or more future time windows.

18. The method of claim 17, further comprising comparing a utilization level of the server during the one or more historical time windows to one or more utilization thresholds, wherein the utilization thresholds correspond to utilization levels between about 1% utilization to about 10% utilization.