APPLICATION EFFICIENCY ENGINE

• US 20100281482A1
• Filed: 04/30/2009
• Published: 11/04/2010
• Est. Priority Date: 04/30/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Fries, Robert, Pike, Robert, Vaid, Kushagra V.
• Granted Patent: US 8,261,266 B2
• Time in Patent Office: Days
• US Class Current: 718/102
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: G06F 9/44505 Configuring for program initiating, e.g. using registry, configuration files
• CPC Class Codes: G06F 9/4856 resumption being on a different machine, e.g. task migration, virtual machine migration G06F9/5088 takes precedence
• CPC Class Codes: G06F 9/5044 considering hardware capabilities

Abstract

A system and a method are provided. Performance and capacity statistics, with respect to an application executing on one or more VMs, may be accessed and collected. The collected performance and capacity statistics may be analyzed to determine an improved hardware profile for efficiently executing the application on a VM. VMs with a virtual hardware configuration matching the improved hardware profile may be scheduled and deployed to execute the application. Performance and capacity statistics, with respect to the VMs, may be periodically analyzed to determine whether a threshold condition has occurred. When the threshold condition has been determined to have occurred, performance and capacity statistics, with respect to VMs having different configurations corresponding to different hardware profiles, may be automatically analyzed to determine an updated improved hardware profile. VMs for executing the application may be redeployed with virtual hardware configurations matching the updated improved profile.

Claims

1. A method for automatically learning an improved hardware profile for an application, the method comprising:

collecting, by a processing device of one or more processing devices of an operating environment, performance statistics of the application executing in a plurality of hardware configurations;

automatically profiling the application, by a processing device of the one or more processing devices of the operating environment, based on the collected performance statistics to derive an improved hardware profile with respect to the application;

scheduling, by a processing device of the one or more processing devices of the operating environment, deployment of a virtual machine on a second processing device of a plurality of second processing devices, the virtual machine having a virtual hardware configuration matching the derived hardware profile; and

deploying the virtual machine, on the second processing device, for executing the application.

2. The method of claim 1, wherein the automatic profiling of the application further comprises:

deriving a role associated with the application based on the stored collected performance statistics, and

associating the role with the hardware profile.

3. The method of claim 2, further comprising:

monitoring, by a processing device of the one or more processing devices of the operating environment, performance statistics of the application executing on the second processing device;

deriving, by a processing device of the one or more processing devices of the operating environment, an updated improved hardware profile based on the monitoring;

performing, by a processing device of the one or more processing devices, the scheduling of a redeployment of the virtual machine, a virtual hardware configuration of the virtual machine matching the derived updated improved hardware profile; and

redeploying the virtual machine on a second processing device of the plurality of second processing devices.

4. The method of claim 3, further comprising:

associating the role with the updated improved hardware profile.

5. The method of claim 3, further comprising:

calculating, by a processing device of the one or more processing devices, processor utilization, amount of memory allocated, number of inputs/outputs per fixed unit of time with respect to a medium, storage space used or available with respect to the medium, and network utilization, with respect to the application executing on the deployed virtual machine, wherein

the deriving of the updated improved hardware profile is based, at least in part, on the processor utilization, the amount of memory allocated, the number of inputs/outputs per fixed unit of time with respect to a medium, the storage space used or available with respect to the medium, and network utilization.

6. The method of claim 1, wherein the scheduling of deployment of a virtual machine further comprises:

comparing the improved hardware profile to available virtual machine resources on the plurality of second processing devices,

determining one of the plurality of second processing devices as having at least available machine resources matching the improved hardware profile, and

scheduling deployment of the virtual machine on the determined one of the plurality of second processing devices.

7. The method of claim 1, further comprising:

distributing, by one or more load balancers, a load for a plurality of copies of the application executing in a plurality of virtual machines with different virtual hardware configurations, the distributing of the load being based on respective response times of the plurality of copies of the application, wherein

the collecting of the performance statistics further include;

collecting application response time data reported by the one or more load balancers.

8. The method of claim 1, further comprising:

analyzing a plurality of hardware profiles with respect to an efficiency achieved by a plurality of virtual machines executing the application and having a virtual hardware configuration matching ones of the plurality of hardware profiles, wherein

the derived improved hardware profile is based, at least in part, on a result of the analyzing.

9. A system for automatically learning an hardware profile for a virtual machine to execute an application at an improved efficiency, the system comprising:

at least one profiler to collect performance and capacity statistics with respect to an application executing in a plurality of hardware configurations;

a trending engine to access the collected performance and capacity statistics and to automatically profile the application to derive a hardware profile and a role of the application, the role of the application being associated with the hardware profile, a virtual machine having a virtual hardware configuration that matches the derived hardware profile being a configuration for executing the application at an improved efficiency;

a scheduler to match the hardware profile to a physical processing device having at least enough available resources for the virtual machine having the virtual hardware configuration that matches the derived hardware profile, the scheduler deploying the virtual machine on the physical processing device.

10. The system of claim 9, further comprising:

at least one load balancer to distribute a load for a plurality of copies of the application executing in a plurality of virtual machines with a plurality of virtual hardware configurations, the at least one load balancer to distribute the load based on application response times of the executing copies of the application.

11. The system of claim 10, wherein:

the at least one load balancer maintains application response time statistics of the copies of the applications executing in the plurality of virtual machines, and

the at least one profiler collects the application response time statistics from the at least one load balancer.

12. The system of claim of claim 9, further comprising:

a monitor for monitoring performance of the application on the deployed virtual machine, wherein

when the monitor detects a performance change beyond a threshold level, with respect to the executing application, the monitor invokes the at least one profiler and the trending engine to automatically derive an updated hardware profile.

13. The system of claim 9, wherein the trending engine derives the hardware profile based, at least in part, on processor utilization, amount of memory allocated, number of inputs/outputs per fixed unit of time with respect to a medium, storage space used or available with respect to the medium, and network utilization with respect to at least one physical processing device including at least one virtual machine executing the application.

14. The system of claim 13, wherein the trending engine further derives the hardware profile based, at least in part, on analyzing a plurality of hardware profiles with respect to an efficiency achieved by respective virtual machines executing the application and having virtual hardware configurations matching ones of the plurality of hardware profiles.

15. A processing device comprising:

at least one processor;

a communications interface for communicating via a network; and

a memory, the communications interface, the at least one processor, and the memory being connected with each other, the memory comprising;

instructions for accessing collected performance and capacity statistics and associated hardware profile information with respect to an application executing on a virtual machine,

instructions for analyzing the collected performance and capacity statistics and the associated hardware profile information and deriving a hardware profile corresponding to a virtual hardware configuration for a virtual machine to execute the application with an improved efficiency, and

instructions for providing information regarding the derived hardware profile to a scheduler for scheduling and deploying a virtual machine having a virtual hardware configuration matching the hardware profile.

16. The processing device of claim 15, wherein the instructions for analyzing the collected performance and capacity statistics and the associated hardware profile information and deriving a hardware profile for a virtual machine to execute the application with an improved efficiency further comprise:

instructions for analyzing a plurality of hardware profiles with respect to an efficiency achieved by a plurality of virtual machines executing the application and having virtual hardware configurations matching ones of the plurality of hardware profiles.

17. The processing device of claim 15, wherein:

the instructions for accessing collected performance and capacity statistics, and associated hardware profile information with respect to an application executing on a virtual machine, further comprise;

instructions for accessing application response time information with respect to the application executing on the virtual machine, and

the instructions for analyzing the collected performance and capacity statistics and the associated hardware profile information and deriving a hardware profile for a virtual machine to execute the application with an improved efficiency further comprise;

instructions for analyzing the application response time information with respect to the application executing on the virtual machine.

18. The processing device of claim 17, wherein the memory further comprises:

instructions for periodically receiving performance and capacity statistics, associated hardware profile information and application response time information, with respect to a deployed virtual machine, while the application is executing on the deployed virtual machine,

instructions for analyzing the periodically received performance and capacity statistics, the associated hardware profile information and the application response time information to determine whether at least one performance threshold condition has occurred, and

instructions for deriving an updated hardware profile when the at least one performance threshold condition has been determined to occur.

19. The processing device of claim 18, wherein a plurality of threshold conditions include a first predefined change in processor utilization lasting at least a first given period of time, a second predefined change in memory allocation lasting at least a second given period of time, a third predefined change in an amount of input/output activity to a medium lasting at least a third given period of time, a fourth predefined change in an amount of network input/output lasting at least a fourth given period of time, and a fifth predefined change in application response time lasting at least a fifth given period of time.

20. The processing device of claim 15, wherein the memory further comprises:

instructions for accessing, from a data repository, performance and capacity statistics, and associated hardware profile information with respect to an application executing on a plurality of virtual machines having a plurality of virtual hardware configurations.