FAILURE SIMULATION AND AVAILABILITY REPORT ON SAME

• US 20090271170A1
• Filed: 04/25/2008
• Published: 10/29/2009
• Est. Priority Date: 04/25/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: De Barros, Marcelo M., Shi, Hui, Mittal, Manish, Forsmann, Joseph
• Granted Patent: US 8,010,325 B2
• Time in Patent Office: Days
• US Class Current: 703/21
• CPC Class Codes: G06F 11/261 by simulating additional hardware, e.g. fault simulation
• 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 11/3457 Performance evaluation by simulation
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 2201/88 Monitoring involving counting

Abstract

A simulation may be performed to assess availability of a service in the event of various types of failures. A specification may describe a behavior to be induced on a machine, and a formula under which availability is to be measured. A controller processes the specification and dispatches a request to induce the behavior to a machine on which the behavior is to be induced. An agent on that machine may look up a routine in a simulation library and may use the routine to induce the behavior. Data, such as performance counter values, are collected before, during, and/or after occurrence of the behavior, and are transmitted back to the controller. A report generator applies the formula contained in the specification to the data received to calculate the availability metric, and produces a report containing that metric.

Claims

1. A method of evaluating failure tolerance, the method comprising:

receiving a specification of a simulation to be performed, said simulation comprising a behavior;

dispatching, to an agent installed at a machine, a request to induce said behavior at said machine;

receiving, from said agent, data collected at said machine; and

generating a report, based on said data, that comprises a first indication of performance during occurrence of said behavior.

2. The method of claim 1, wherein said data comprises values collected from a plurality of performance counters, and wherein said generating comprises:

obtaining, from said specification, a formula that involves said performance counters; and

calculating a result based on values of said performance counters, said report comprising said result.

3. The method of claim 1, wherein said generating comprises:

including, in said report, a second indication of performance prior to occurrence of said behavior, and a third indication of performance after occurrence of said behavior.

4. The method of claim 1, further comprising:

displaying said report during occurrence of said behavior; and

updating said report based on new data collected at said machine during occurrence of said behavior.

5. The method of claim 1, wherein said machine is, or is part of, a system that processes transactions, and wherein said first indication of performance describes throughput of said transactions through said system.

6. The method of claim 1, wherein said machine is, or is part of, a system that provides a service, and wherein said first indication of performance describes availability of said service.

7. One or more computer-readable storage media comprising executable instructions to perform a method, the method comprising:

receiving, from a controller that operates at a first machine, a request to induce a behavior at a second machine on which said instructions execute;

recording a first value of a performance counter prior to said behavior;

inducing said behavior in response to said request;

recording a second value of said performance counter prior to said behavior and a third value of said performance counter after said behavior has ended;

communicating, to said controller, data that comprises said first value, said second value, and said third value.

8. The one or more computer-readable storage media of claim 7, wherein said behavior comprises non-operation of a program, and wherein said inducing comprises:

stopping said program.

9. The one or more computer-readable storage media of claim 7, wherein said behavior comprises using a processor up to a percentage of its capacity, and wherein said inducing comprises:

causing code to execute on said processor until utilization of said processor exceeds said percentage.

10. The one or more computer-readable storage media of claim 7, wherein said behavior comprises unavailability of a data storage component, and wherein said inducing comprises:

writing to said data storage component until said data storage component has reached capacity.

11. The one or more computer-readable storage media of claim 7, wherein said behavior interferes with said controller'"'"'s ability to communicate with said second machine, and wherein the method further comprises:

ceasing said behavior without a contemporaneous direction from said controller cease said behavior.

12. The one or more computer-readable storage media of claim 7, wherein the method further comprises:

waiting for an event to finish prior to said inducing of said behavior.

13. The one or more computer-readable storage media of claim 12, wherein the method further comprises:

reporting a time at which said behavior commences, said time being after completion of said event.

14. The one or more computer-readable storage media of claim 7, wherein the method further comprises:

reporting a time at which said behavior terminates.

15. A system comprising:

a first computer on which a controller is installed;

a first instance of an agent that induces, on a second computer on which said first instance of said agent is installed, a first behavior in response to a first request from said controller, and that collects data about performance on said second computer;

a report generator that receives said data and that generates, based on information that comprises said data, a report that describes availability of a service provided by a group of one or more computers that include said second computer; and

an output device that is comprised by, or connected to, said first computer, said report generator communicating said report through said output device.

16. The system of claim 15, further comprising:

a library that is accessible by said first instance of said agent and that contains a program that, when executed, causes said second computer to carry out said first behavior, said first instance of said agent inducing said first behavior by retrieving said program from said library and causing said program to execute.

17. The system of claim 15, further comprising:

a second instance of an agent installed on a third computer, wherein said first request is that said first instance of said agent induce said first behavior at a first time, and wherein said controller issues, to said second instance of said agent, a second request that said second instance of said agent induce a second behavior at a second time that differs from said first time.

18. The system of claim 15, further comprising:

a second instance of an agent that induces, on a third computer on which said second instance of said agent is installed, a second behavior in response to a second request from said controller, wherein said information on which generation of said report is based further comprises performance at said third computer.

19. The system of claim 15, wherein said controller receives a specification that comprises a formula based a performance counter, said data comprising a value of said performance counter, wherein said report generator applies said formula to said value to generate a result, and includes said result in said report.

20. The system of claim 15, wherein said report describes availability of said service before occurrence of said first behavior, during occurrence of said first behavior, and after occurrence of said first behavior.