Predicting performance of a consolidated virtualized computing environment

• US 9,164,785 B2
• Filed: 03/13/2012
• Issued: 10/20/2015
• Est. Priority Date: 03/14/2011
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Kraft, Stephan, Krishnamurthy, Diwakar, Casale, Giuliano
• Primary Examiner(s): Rivas, Omar Fernandez
• Assistant Examiner(s): KHAN, IFTEKHAR A
• Application Number:
  US13/418,879
• Publication Number:
  US 20120239376A1
• Time in Patent Office: 1,316 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/301 where the computing system is a virtual computing platform, e.g. logically partitioned systems virtual machines G06F9/45533; logical partitioning of resources G06F9/5077
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• 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 9/45533 Hypervisors; Virtual machine monitors

Abstract

Implementations of the present disclosure provide computer-implemented methods for predicting a performance of a consolidated virtualized computing environment. Methods include processing benchmark workloads using a plurality of virtual machines to generate a plurality of traces, each trace including data corresponding to requests issued by a respective virtual machine operating in an isolated environment, storing the plurality of traces in one or more trace repositories, each trace repository provided as a computer-readable storage medium, selecting a trace from the plurality of traces stored in the one or more trace repositories, parameterizing a queuing model based on the trace, the queuing model representing request queuing in the consolidated virtualized computing environment, and processing the queuing model using one or more processors to generate one or more response time estimates for the consolidated virtualized computing environment.

Claims

1. A computer-implemented method of predicting a performance of a consolidated virtualized computing environment, comprising:

generating, by one or more processors, a plurality of traces by processing benchmark workloads using a plurality of virtual machines, each trace being collected inside a respective virtual machine and comprising data corresponding to requests issued by the respective virtual machine operating in an isolated environment, the data comprising a block size of each request;

storing, by one or more processors, the plurality of traces in one or more trace repositories, each trace repository provided as a computer-readable storage medium;

selecting, by one or more processors, a trace from the plurality of traces stored in the one or more trace repositories;

performing, by the one or more processors, one or more pre-processing operations on the trace, the one or more pre-processing operations comprising splitting at least one request comprising a respective block size that exceeds a threshold block size, a total amount of splitting the at least one request being determined from a ceiling function applied to a remainder of a division of the at least one request based on the threshold block size;

parameterizing, by one or more processors, a queuing model based on the trace and the one or more pre-processing operations, the queuing model representing request queuing in the consolidated virtualized computing environment; and

processing the queuing model using one or more processors to generate one or more response time estimates for the consolidated virtualized computing environment.

2. The method of claim 1, wherein a first isolated environment comprises a first virtual machine and a virtual machine monitor (VMM), a second isolated environment comprises a second virtual machine and the VMM, and the consolidated virtualized computing environment comprises the first virtual machine, the second virtual machine and the VMM.

3. The method of claim 1, wherein the plurality of traces comprise request arrival traces, each request arrival trace being associated with a respective isolated environment.

4. The method of claim 1, wherein the queuing model comprises a merge value parameter that approximates a number of merged requests.

5. The method of claim 4, wherein the merge value parameter is determined based on the trace.

6. The method of claim 4, wherein parameterizing the queuing model comprises executing an iterative search to estimate the merge value parameter.

7. The method of claim 6, wherein executing the iterative search comprises:

determining a plurality of expected mean queue lengths and a plurality of effective mean queue lengths;

determining a merge error based on an expected mean queue length of the plurality of expected mean queue lengths, an effective mean queue length of the plurality of effective mean queue lengths, and an estimated merge value parameter; and

determining that the merge error is less than a threshold merge error and, in response, selecting the estimated merge value parameter as the merge value parameter.

8. The method of claim 7, wherein the effective mean queue length is determined based on the estimated merge value parameter.

9. The method of claim 1, wherein the queuing model is provided as a multiclass open queuing model that distinguishes requests submitted from individual virtual machines in separate classes.

10. The method of claim 1, wherein the queuing model operates using start-time fair queuing (SFQ) scheduling.

11. The method of claim 1, wherein each trace is generated based on a completely fair queuing (CFQ) scheduler of the respective virtual machine, the CFQ scheduler having internal queues to manage input and output (I/O) requests.

12. The method of claim 1, wherein the consolidated virtualized computing environment comprises two or more virtual machines operating on a virtualized server.

13. A non-transitory computer-readable storage medium coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for improving keyword searches, the operations comprising:

generating a plurality of traces by processing benchmark workloads using a plurality of virtual machines, each trace being collected inside a respective virtual machine and comprising data corresponding to requests issued by the respective virtual machine operating in an isolated environment, the data comprising a block size of each request;

storing the plurality of traces in one or more trace repositories, each trace repository provided as a computer-readable storage medium;

selecting a trace from the plurality of traces stored in the one or more trace repositories;

performing one or more pre-processing operations on the trace, the one or more pre-processing operations comprising splitting at least one request comprising a respective block size that exceeds a threshold block size, a total amount of splitting the at least one request being determined from a ceiling function applied to a remainder of a division of the at least one request based on the threshold block size;

parameterizing a queuing model based on the trace and the one or more pre-processing operations, the queuing model representing request queuing in a consolidated virtualized computing environment; and

processing the queuing model using one or more processors to generate one or more response time estimates for the consolidated virtualized computing environment.

14. A system, comprising:

a computing device; and

a computer-readable storage device coupled to the computing device and having instructions stored thereon which, when executed by the computing device, cause the computing device to perform operations for improving keyword searches for enterprise services, the operations comprising;

generating a plurality of traces by processing benchmark workloads using a plurality of virtual machines, each trace being collected inside a respective virtual machine and comprising data corresponding to requests issued by the respective virtual machine operating in an isolated environment, the data comprising a block size of each request;

storing the plurality of traces in one or more trace repositories, each trace repository provided as a computer-readable storage medium;

selecting a trace from the plurality of traces stored in the one or more trace repositories;

performing one or more pre-processing operations on the trace, the one or more pre-processing operations comprising splitting at least one request comprising a respective block size that exceeds a threshold block size, a total amount of splitting the at least one request being determined from a ceiling function applied to a remainder of a division of the at least one request based on the threshold block size;

parameterizing a queuing model based on the trace and the one or more pre-processing operations, the queuing model representing request queuing in a consolidated virtualized computing environment; and

processing the queuing model using one or more processors to generate one or more response time estimates for the consolidated virtualized computing environment.