Storage input/output utilization associated with a software application

• US 8,527,238 B2
• Filed: 09/30/2008
• Issued: 09/03/2013
• Est. Priority Date: 09/30/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Hollingsworth, Matt, Oslake, John M.
• Primary Examiner(s): CHERRY, STEPHEN J
• Application Number:
  US12/242,710
• Publication Number:
  US 20100082300A1
• Time in Patent Office: 1,799 Days
• Field of Search: 702/186
• US Class Current: 702/186
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A method includes determining a total storage input/output capacity of a set of storage elements available for use by a software application located on an operating system volume of a computer system. The total storage input/output capacity is determined by evaluating storage input/output capacity of the storage elements at a physical device level and at a file level. The method includes measuring a number of input/output storage operations performed on behalf of the software application. The method also includes determining a storage input/output utilization of the software application based on the measured number of input/output storage operations as compared to the total storage input/output capacity of the set of storage elements.

Claims

1. A method comprising:

identifying a set of storage elements mapped to a software application;

determining a total storage input/output capacity of the set of storage elements mapped to the software application located on an operating system volume of a computer system, wherein the total storage input/output capacity is determined by a processor evaluating storage input/output capacity of the storage elements at a physical device level, at an intermediate level, and at a file level;

measuring a number of input/output storage operations performed on behalf of the software application; and

determining a storage input/output utilization of the software application at the processor based on the measured number of input/output storage operations as compared to the total storage input/output capacity of the set of storage elements.

2. The method of claim 1, wherein the set of storage elements is identified as mapped to the software application by use of a computing system model that identifies the set of storage elements and wherein the computing system model includes a model of components of a computing system and their relationships at a file level, at an operating system volume level, at an operating system disk level, and at a physical storage level, and wherein each of the components of the computer system model is available to satisfy a storage input/output request of the software application, and further comprising determining a resource governed utilization of the storage input/output capacity based on historical usage of the storage input/output capacity as compared to a governed storage input/output capacity available and a non-governed storage input/output capacity available.

3. The method of claim 1, further comprising:

determining the total storage input/output utilization as a ratio of the measured number of input/output storage operations and the total storage input/output capacity; and

displaying the storage input/output utilization of the software application.

4. The method of claim 1, wherein the set of storage elements comprises a plurality of physical storage devices, the method further comprising identifying one of the storage elements of the set of storage elements as a potential bottleneck in an input/output path, and wherein the software application is a structured query language database file that is physically constrained to a specific number of input/output storage accesses per second.

5. The method of claim 4, wherein an operating system level input/output request traverses one or more alternative paths to reach a target, wherein the target includes one or more of the physical storage devices.

6. The method of claim 1, further comprising recording a workload of the software application and replaying the recorded workload at a computing system with an increased rate or an increased concurrency.

7. The method of claim 1, further comprising using a combination of different storage instrumentation application programming interfaces to determine storage input/output utilization of each of the storage elements in the set of storage elements.

8. The method of claim 1, further comprising detecting, based on the storage input/output utilization, a usage of the software application that approaches a storage input/output resource limit before the usage exceeds the storage input/output resource limit.

9. The method of claim 1, wherein evaluating storage input/output capacity comprises:

determining characteristics of each storage element at the physical storage device level, each storage element including physical storage devices or portions of shared physical storage devices;

collecting metrics of each storage element at an operating system file level; and

determining the total storage input/output capacity based on the characteristics and based on the metrics.

10. The method of claim 1, wherein the storage capacity of each storage element in the set of storage elements corresponds to the total storage input/output capacity accessible to the software application, and wherein the set of storage elements are a subset of components of the computer system and wherein the set of storage elements are those components of the computer system that are available for use by the software application.

11. A method, comprising:

relating an operating system file to a subset of a plurality of physical storage devices of a computer system, wherein the subset of the plurality of physical storage devices indicates those storage devices that are in an input/output path of the operating system file;

sending a query to application programming interfaces associated with the subset of the plurality of physical storage devices to determine one or more device properties, configuration data, and one or more usage statistics;

receiving responses from the application programming interfaces;

calculating a total available storage, input/output capacity of the operating system file to based on the responses;

determining storage input/output usage of the operating system file; and

determining storage input/output utilization of the operating system file based on the storage input/output usage of the operating system file and based on the total available storage input/output capacity of the operating system file.

12. The method of claim 11, further comprising:

modeling components of the computing system that include the plurality of physical storage devices;

modeling relationships between the components; and

determining the total available storage input/output capacity of the operating system file based on a given workload and based on a storage input/output capacity of those components of the computing system identified as available for use by the operating system file and within the subset of the plurality of storage devices, wherein the application programming interfaces (APIs) include a first API of a first API type and a second API of a second API type, and wherein the second API is a storage instrumentation API.

13. The method of claim 11, wherein calculating the total available storage input/output capacity of the operating system file includes:

computing an input/output capacity of each of multiple storage device elements in an input/output path of the operating system file;

identifying the storage device element of the multiple storage device elements that has a minimum storage input/output capacity; and

wherein the identified storage device element that has the minimum storage input/output capacity represents the total available storage input/output capacity of the operating system file.

14. An apparatus comprising a computer readable storage device comprising instructions that executable by a processor to:

retrieve system model data stored in a memory device, the system model data identifying components of a modeled system and including information describing relationships between the components, wherein each of the components is associated with performing an input/output storage operation of a software application, the software application mapped to each of the components by a data structure of the system model data stored in the memory device;

call application programming interfaces for each of the components and to receive storage input/output capacity data for each of the components;

collect the received storage input/output capacity for each of the components to determine a total available storage input/output capacity for the software application;

retrieve file storage input/output usage data associated with the software application, the file storage input/output usage data retrieved by periodically querying a database platform to obtain measured usage data of the components;

calculate a storage input/output utilization associated with the software application based on the retrieved file storage input/output usage data and the total available storage input/output capacity for the software application; and

present the storage input/output utilization to a display device.

15. The apparatus of claim 14, wherein the application programming interfaces comprise at least one of a vendor application programming interface (API), a Storage Management Initiative (SMI) provider API, a common information management object manager (CIMOM) API, a common information management extensible markup language (CIM-XML) API, a web services for management (WS-Man) API, a system services application API, and a performance metric API.

16. The apparatus of claim 14, wherein a recorded workload is applied to a target component of the modeled system at an increased rate or an increased concurrency until a measured rate of increase of input/output latency becomes non-linear.

17. The apparatus of claim 16, wherein a measured storage capacity of the target component at the time when the rate of increase of the input/output latency becomes nonlinear is the storage input/output storage capacity of the target component that is used to determine the total available input/output capacity for the software application.

18. The apparatus of claim 14, wherein the received storage input/output capacity for each of the components is collected via concurrent paths or via alternative paths.

19. The apparatus of claim 14, wherein the display device displays capacity viewpoints indicating input/output utilization of the software application at a physical storage level, at an operating system disk level, at a volume level, and at a file level.

20. The apparatus of claim 14, wherein an allocated space utilization region in a view displayed on the display device displays a percentage of allocated space utilized by the software application when the software application is limited from using a total space available for a volume by an allocation restriction of another software application.