Determining a device wear-rate

Patent metadata

Field	Value
Publication number	US-10860218-B2
Application number	US-201616091050-A
Country	US
Kind code	B2
Filing date	Jul 8, 2016
Priority date	Jul 8, 2016
Publication date	Dec 8, 2020
Grant date	Dec 8, 2020

Abstract

Example implementations relate to determining a device wear-rate. An example system for determining a device wear-rate can include a plurality of filter drivers to: monitor system requests for I/O associated with a device of the system and transmit information associated with the system requests to a filter manager. The system can also include the filter manager to catalog the information, a service to collate the information across a plurality of machine configurations and workloads, and a processor to determine a wear-rate of the device based on an analysis of the collated information.

First claim

What is claimed:

1. A system, comprising:
a plurality of filter drivers to:
monitor system requests for input/output (I/O) associated with a device of the system; and
transmit information associated with the system requests to a filter manager;

the filter manager to:
tag the information;
catalog the tagged information; and
compare the tagged information to a data payload of the I/O requests exercised against physical segments of the device;

a service to collate the information across a plurality of machine configurations and workloads; and
a processor to determine a wear-rate of the physical segments of the device based on an analysis of the collated information and comparison of the tagged information.

2. The system of claim 1 , further comprising the plurality of filter drivers to monitor system requests for I/O from direct application programming interfaces.

3. The system of claim 1 , further comprising the plurality of filter drivers to monitor system requests for I/O from memory-mapped I/Os.

4. The system of claim 1 , further comprising the filter manager to catalog the system requests based on application context, amount of I/O, type of I/O, and time.

5. The system of claim 1 , further comprising the processor to analyze the collated information to determine groupings of commonality and determine an outlier within the system.

6. The system of claim 1 , further comprising the processor to determine when the wear-rate of the device surpasses a particular threshold.

7. A method, comprising,
collecting, using a file system device driver, information associated with a plurality of applications associated with a system by analyzing behavior of the applications;
tagging the information associated with the plurality of applications;
comparing the tagged information to a data payload of I/O requests exercised against physical segments of a plurality of devices;
correlating the collected information in a plurality of concentrations;
determining outliers in each of the plurality of concentrations;
correlating the outliers with other outliers exhibiting similar behavior within the system;
determining failure probabilities for the plurality of devices in the system based on the correlation; and
determining a wear-rate of one of the plurality of devices based on the failure probabilities of the physical segments of the one of the plurality of devices.

8. The method of claim 7 , wherein collecting information associated with the plurality of applications comprises analyzing read and write activities for each of the plurality of applications.

9. The method of claim 8 , wherein analyzing read and write activities comprises determining lengths of read and write activities of each of the plurality of applications.

10. The method of claim 7 , further comprising comparing performances of each of the plurality of devices to one another based on an application load associated with each of the plurality of devices and an amount of time the application load was applied to each of the plurality of devices.

11. The method of claim 7 , further comprising creating a filtering subsystem of the system based on the wear-rate determination.

12. The method of claim 7 , wherein determining outliers comprises determining system devices within each of the plurality of concentrations having faster wear-rates as compared to other system devices within each of the plurality of concentrations.

13. A non-transitory machine-readable medium storing instructions executable by a processing resource to cause a computing system to:
monitor interactions between a file system and an application layer of the computing system;
tag the interactions;
compare the tagged information to a data payload of the interactions exercised against physical segments of a plurality of system devices;
determine, using a file system filter layer, a plurality of applications of the application layer that are reading and writing data;
determine how much data each of the plurality of applications is reading and writing to the physical segments of the plurality of system devices;
create system device usage patterns based on the monitored interactions, determined plurality of applications, data payload of the interactions against the physical segments, and determined amount of data;
collate the usage patterns across the plurality of system devices using a network services layer and a database; and
determine wear-rate patterns of the plurality of system devices based on the collation.

14. The non-transitory machine-readable medium of claim 13 , further comprising the instructions executable to determine from which files associated with the computing system the plurality of applications are reading and writing.

15. The non-transitory machine-readable medium of claim 13 , wherein the instructions executable to collate the usage patterns comprising instructions executable to collate the usage patterns based on at least one of a location of one of the plurality of system devices, a usage date of one of the plurality of system devices, a usage time of one of the plurality of system devices, a business group of one of the plurality of system devices, and application usage of one of the plurality of system devices.

Assignees

• Hewlett Packard Development Co

Inventors

• Graham Christoph J
• Flynn Thomas J
• Herrera Virginia Quance

Classifications

• G06F11/3072 Primary

  where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting · CPC title

• G06F11/3485

  for I/O devices · CPC title

• G06F11/008

  Reliability or availability analysis · CPC title

• G06F11/004

  Error avoidance (G06F11/07 and subgroups take precedence) · CPC title

• G06F11/3058

  Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations (thermal management in cooling arrangements of a computing system G06F1/206) · CPC title