Management of heterogeneous workloads

• US 9,405,585 B2
• Filed: 04/30/2007
• Issued: 08/02/2016
• Est. Priority Date: 04/30/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chess, David Michael, Steinder, Malgorzata, Whalley, Ian Nicholas
• Primary Examiner(s): Miller, Alan S
• Assistant Examiner(s): Swartz, Stephen S
• Application Number:
  US11/741,875
• Publication Number:
  US 20080270199A1
• Time in Patent Office: 3,382 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 2209/5017 Task decomposition
• CPC Class Codes: G06F 9/455 Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06Q 10/06316 Sequencing of tasks or work

Abstract

Systems and methods for managing a system of heterogeneous workloads are provided. Work that enters the system is separated into a plurality of heterogeneous workloads. A plurality of high-level quality of service goals is gathered. At least one of the plurality of high-level quality of service goals corresponds to each of the plurality of heterogeneous workloads. A plurality of control functions are determined that are provided by virtualizations on one or more containers in which one or more of the plurality of heterogeneous workloads run. An expected utility of a plurality of settings of at least one of the plurality of control functions is determined in response to the plurality of high-level quality of service goals. At least one of the plurality of control functions is exercised in response to the expected utility to effect changes in the behavior of the system.

Claims

1. A method for managing a system of heterogeneous workloads, comprising the steps of:

separating work that enters the system into a plurality of heterogeneous workloads, wherein the step of separating work comprises the steps of;

receiving web application requests at a request router; and

receiving jobs at a job scheduler;

gathering a plurality of high-level quality of service goals, wherein at least one of the plurality of high-level quality of service goals corresponds to each of the plurality of heterogeneous workloads, wherein the high-level quality of service goals comprise information regarding at least one of a target application and a service class of the heterogeneous workloads;

determining a plurality of control functions that are provided by virtualizations on one or more containers in which one or more of the plurality of heterogeneous workloads run, wherein the step of determining a plurality of control functions comprises the steps of;

placing web applications on one or more of a plurality of heterogeneous server machines;

dispatching the web application requests from the request router to one or more web applications on one or more of the plurality of heterogeneous server machines in accordance with a scheduling mechanism; and

managing virtual machines inside a given heterogeneous server machine by a machine agent in a first domain of the given heterogeneous server machine in communication with a node agent in a second domain of the given heterogeneous server machine, wherein;

the given heterogeneous server machine is a node including each of the machine agent, node agent, first domain and second domain therein;

the machine agent includes a command executor operatively coupled to the node agent, and the node agent interacts with the command executor of the machine agent to control a job; and

the command executor is further operatively coupled to an image management subsystem which dispenses pre-created virtual images for use during runtime, and a configuration management subsystem which provides the pre-created virtual images with configuration files prior to booting to set up devices and networking for the pre-created virtual images;

wherein a process is created in at least one of the pre-created virtual images by populating the pre-created virtual image with process data necessary to run the process; and

wherein, after the pre-created virtual image is used, the used pre-created virtual image is scrubbed of the process data so that the pre-created virtual image can be reused by a future process;

determining an expected utility of a plurality of settings of at least one of the plurality of control functions in response to the plurality of high-level quality of service goals;

exercising at least one of the plurality of control functions in response to the expected utility to effect changes in the behavior of the system, wherein the step of exercising at least one of the one or more control functions comprises the step of dynamically adjusting the scheduling mechanism in response to changing workload intensity and system configuration, and wherein the step of dynamically adjusting the scheduling mechanism comprises the steps of;

building a model of the system that allows a flow controller to predict performance of a flow for any choice of concurrency limit and weights of a weighted-fair scheduling discipline; and

dynamically adjusting the weights of the weighted-fair scheduling discipline and the concurrency limit of the system in response to the changing workload intensity and system configuration;

at least one of suspending or pausing a given pre-created virtual image that is running on the given heterogeneous server machine;

transferring and copying the suspended or paused pre-created virtual image from the given heterogeneous server machine to another node; and

restoring or resuming the suspended or paused pre-created virtual image on the other node;

wherein the steps of separating, gathering, determining a plurality of control functions, determining an expected utility, exercising at least one of the plurality of control functions, suspending or pausing, transferring and copying, and restoring or resuming are executed by a processor device.

2. The method of claim 1, wherein the placing is performed through a placement controller driven by utility functions of allocated CPU demand, and the step of determining a plurality of control functions further comprises the step of:

attributing jobs to one or more of the plurality of heterogeneous server machines in accordance with placement decisions communicated to a job scheduler by the placement controller.

3. The method of claim 2, wherein the step of placing web applications further comprises the step of calculating a web application placement that maximizes the minimum utility across all web applications.

4. The method of claim 2, wherein the step of dispatching the web application requests comprises the steps of:

classifying requests into flows at the request router depending on at least one of target web application and service class;

placing classified requests into per-flow queues within the request router; and

dispatching requests from the per-flow queues based on a weighted-fair scheduling discipline that observes a concurrency limit of the system.

5. The method of claim 2, wherein the step of attributing jobs comprises the steps of:

managing dependencies among jobs at the job scheduler;

matching jobs with one or more of the plurality of heterogeneous server machines;

receiving jobs from the job scheduler at a job scheduler proxy of the placement controller;

estimating requirements for jobs for a given value of utility function at a job utility estimator of the placement controller;

calculating job placement that maximizes the minimum utility across all web applications at a placement optimizer of the placement controller;

communicating the job placements to the job scheduler through the job scheduler proxy; and

attributing jobs to one or more of the plurality of heterogeneous server machines in accordance with the communicated job placements.

6. The method of claim 2, wherein the step of determining one or more control functions further comprises the step of:

providing job management functionality within the heterogeneous server machine through interaction with the machine agent.

7. The method of claim 2, wherein the step of determining an expected utility comprises the steps of:

obtaining profiles for web application requests in the form of an average number of CPU cycles consumed by requests of a given flow;

obtaining profiles for jobs in the form of at least one of the number of CPU cycles required to complete the job, the number of threads used by the job, and the maximum CPU speed at which the job may progress; and

communicating the profiles for web application requests to the placement controller and the flow controller, and the profiles for jobs to the job scheduler.

8. Apparatus for managing a system of heterogeneous workloads, comprising:

a memory; and

at least one processor coupled to the memory and operative to;

(i) separate work that enters the system into a plurality of heterogeneous workloads, wherein in separating the work, the at least one processor is further operative to;

receive web application requests at a request router; and

receive jobs at a job scheduler;

(ii) gather a plurality of high-level quality of service goals, wherein at least one of the plurality of high-level quality of service goals corresponds to each of the plurality of heterogeneous workloads, wherein the high-level quality of service goals comprise information regarding at least one of a target application and a service class of the heterogeneous workloads;

(iii) determine a plurality of control functions that are provided by virtualizations on one or more containers in which one or more of the plurality of heterogeneous workloads run, wherein in determining the plurality of control functions, the at least one processor is further operative to;

place web applications on one or more of a plurality of heterogeneous server machines;

dispatch the web application requests from the request router to one or more web applications on one or more of the plurality of heterogeneous server machines in accordance with a scheduling mechanism; and

manage virtual machines inside a given heterogeneous server machine by a machine agent in a first domain of the given heterogeneous server machine in communication with a node agent in a second domain of the given heterogeneous server machine, wherein;

the given heterogeneous server machine is a node including each of the machine agent, node agent, first domain and second domain therein;

the machine agent includes a command executor operatively coupled to the node agent, and the node agent interacts with the command executor of the machine agent to control a job; and

the command executor is further operatively coupled to an image management subsystem which dispenses pre-created virtual images for use during runtime, and a configuration management subsystem which provides the pre-created virtual images with configuration files prior to booting to set up devices and networking for the pre-created virtual images;

wherein a process is created in at least one of the pre-created virtual images by populating the pre-created virtual image with process data necessary to run the process; and

wherein, after the pre-created virtual image is used, the used pre-created virtual image is scrubbed of the process data so that the pre-created virtual image can be reused by a future process;

(iv) determine an expected utility of a plurality of settings of at least one of the plurality of control functions in response to the plurality of high-level quality of service goals;

(v) exercise at least one of the plurality of control functions in response to the expected utility to effect changes in the behavior of the system, wherein in exercising at least one of the one or more control functions, the at least one processor is further operative to dynamically adjust the scheduling mechanism in response to changing workload intensity and system configuration, and wherein in dynamically adjusting the scheduling mechanism, the at least one processor is further operative to;

build a model of the system that allows a flow controller to predict performance of a flow for any choice of concurrency limit and weights of a weighted-fair scheduling discipline; and

dynamically adjust the weights of the weighted-fair scheduling discipline and the concurrency limit of the system in response to the changing workload intensity and system configuration;

(vi) at least one of suspend or pause a given pre-created virtual image that is running on the given heterogeneous server machine;

(vii) transfer and copy the suspended or paused pre-created virtual image from the given heterogeneous server machine to another node; and

(viii) restore or resume the suspended or paused pre-created virtual image on the other node.

9. A method for making a computer implemented process to enable management of a system of heterogeneous workloads comprising the steps of:

instantiating first computer instructions onto a computer readable medium, the first computer instructions configured to separate work that enters the system into a plurality of heterogeneous workloads, wherein separating the work comprises;

receiving web application requests at a request router; and

receiving jobs at a job scheduler;

instantiating second computer instructions onto a computer readable medium, the second computer instructions configured to gather a plurality of high-level quality of service goals, wherein at least one of the plurality of high-level quality of service goals corresponds to each of the plurality of heterogeneous workloads, wherein the high-level quality of service goals comprise information regarding at least one of a target application and a service class of the heterogeneous workloads;

instantiating third computer instructions onto a computer readable medium, the third computer instructions configured to determine a plurality of control functions that are provided by virtualizations on one or more containers in which one or more of the plurality of heterogeneous workloads run, wherein determining the plurality of control functions comprises;

placing web applications on one or more of a plurality of heterogeneous server machines;

dispatching the web application requests from the request router to one or more web applications on one or more of the plurality of heterogeneous server machines in accordance with a scheduling mechanism; and

managing virtual machines inside a given heterogeneous server machine by a machine agent in a first domain of the given heterogeneous server machine in communication with a node agent in a second domain of the given heterogeneous server machine, wherein;

the given heterogeneous server machine is a node including each of the machine agent, node agent, first domain and second domain therein;

the machine agent includes a command executor operatively coupled to the node agent, and the node agent interacts with the command executor of the machine agent to control a job; and

the command executor is further operatively coupled to an image management subsystem which dispenses pre-created virtual images for use during runtime, and a configuration management subsystem which provides the pre-created virtual images with configuration files prior to booting to set up devices and networking for the pre-created virtual images;

wherein a process is created in at least one of the pre-created virtual images by populating the pre-created virtual image with process data necessary to run the process; and

wherein, after the pre-created virtual image is used, the used pre-created virtual image is scrubbed of the process data so that the pre-created virtual image can be reused by a future process;

instantiating fourth computer instructions onto a computer readable medium, the fourth computer instructions configured to determine an expected utility of a plurality of settings of at least one of the plurality of control functions in response to the plurality of high-level quality of service goals;

instantiating fifth computer instructions onto a computer readable medium, the fifth computer instructions configured to exercise at least one of the plurality of control functions in response to the expected utility to effect changes in the behavior of the system, wherein exercising at least one of the one or more control functions comprises dynamically adjusting the scheduling mechanism in response to changing workload intensity and system configuration, and wherein dynamically adjusting the scheduling mechanism comprises;

building a model of the system that allows a flow controller to predict performance of a flow for any choice of concurrency limit and weights of a weighted-fair scheduling discipline; and

dynamically adjusting the weights of the weighted-fair scheduling discipline and the concurrency limit of the system in response to the changing workload intensity and system configuration;

instantiating sixth computer instructions onto a computer readable medium, the sixth computer instructions configured to at least one of suspend or pause a given pre-created virtual image that is running on the given heterogeneous server machine;

instantiating seventh computer instructions onto a computer readable medium, the seventh computer instructions configured to transfer and copy the suspended or paused pre-created virtual image from the given heterogeneous server machine to another node; and

instantiating eighth computer instructions onto a computer readable medium, the eighth computer instructions configured to restore or resume the suspended or paused pre-created virtual image on the other node.