Method and apparatus for policy-based provisioning in a virtualized service delivery environment

• US 8,028,048 B2
• Filed: 02/27/2007
• Issued: 09/27/2011
• Est. Priority Date: 02/27/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation, International Business Machines SA (International Business Machines Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Mohindra, Ajay, Segal, Alla, Karve, Alexei A.
• Primary Examiner(s): Chea; Philip
• Assistant Examiner(s): Shepperd; Eric W
• Application Number:
  US11/679,223
• Publication Number:
  US 20080209016A1
• Time in Patent Office: 1,673 Days
• Field of Search: 709/225, 709/222, 709/221, 709/224, 717/177, 717/178, 702/188, 718/107
• US Class Current: 709/221
• CPC Class Codes: G06F 2009/45562 Creating, deleting, cloning virtual machine instances
• CPC Class Codes: G06F 8/61 Installation
• CPC Class Codes: G06F 9/44505 Configuring for program initiating, e.g. using registry, configuration files
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/0806 for initial configuration or provisioning, e.g. plug-and-play
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0894 Policy-based network configuration management
• CPC Class Codes: H04L 41/0895 Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
• CPC Class Codes: H04L 67/34 involving the movement of software or configuration parameters network booting or remote initial program loading [RIPL] G06F9/4416
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

In a distributed computing network having computers interconnected with each other, a method and apparatus of providing policy based provisioning of middleware components to provide a solution for a service. With this invention, the middleware components that are necessary to provide the solution are determined. Then, the configurations of the computers are determined so as to satisfy a selected configuration policy for a selected solution. Finally, the necessary middleware components are installed on the configured computers so that the solution can be provided. Some of the selected configuration policies are: maximization of server utilization policy, provisioning of software components requiring external access based on availability of external resources, and minimization of total time to install said determined components. The progress of installation of the middleware components can also be monitored to make any necessary adjustments in the configuration of the virtual machines so that the selected configuration policy can be satisfied.

Claims

1. In a distributed computing network having a plurality of computers interconnected with each other and a storage device, a method of providing policy based provisioning of middleware components to provide a solution, said method comprising:

determining a selection of said middleware components that are necessary to provide said solution from said storage device;

determining a predictive time for each of said selected middleware components to be installed on respective ones of each of a number of said computers;

installing each of said selected middleware components on said number of computers so that each said selected middleware component installation is completed at approximately the same time;

monitoring the progress of installation of each of said selected middleware components during said installing;

computing a deviation of the predictive time from an actual time based on said monitoring; and

dynamically adjusting the configuration of said number of said computers during said installing by adjusting at least one of the following parameters of a virtual machine of each of said number of said computers;

memory size, disk size, and percent usage of a central processing unit so that each said selected middleware component installation is completed at approximately the same time.

2. A method according to claim 1, further comprising:

computing an actual time to install said middleware components; and

receiving events or monitoring logs from different progress monitors.

3. A program storage device readable by a plurality of computers interconnected with each other in a distributed computing network, said device having a program of instructions which are tangibly embodied on said storage device and which are executable by said computers, to perform a method of providing policy based provisioning of middleware components to provide a solution, said method comprising:

determining a selection of said middleware components that are necessary to provide said solution;

determining the configuration of each of a number of said computers on which said selected middleware components are to deployed so as to satisfy a selected configuration policy;

installing said selected middleware components on said number of said computers;

monitoring the progress of installation of each of said selected middleware components during said installing; and

computing a deviation of a predictive value associated with said selected configuration policy from an actual value based on said monitoring;

recomputing an objective function of said selected configuration policy; and

dynamically adjusting the configuration of said number of said computers during said installing by adjusting at least one of the following parameters of a virtual machine of each of said number of said computers;

memory size, disk size, and percent usage of a central processing unit to satisfy said selected configuration policy.

4. A method as recited in claim 3, wherein said selected configuration policy comprises at least one of the following:

maximization of server utilization for servers on which virtual machines are setup, provisioning of software components requiring external access based on availability of external resources, and minimization of total time to install said determined components.

5. A method of providing policy based provisioning of middleware components to provide a solution, said method comprising:

determining middleware components that are necessary to provide a solution by accessing a storage device;

determining an initial configuration of virtual machines for a plurality of computers on which the middleware components are to deployed to satisfy a selected configuration policy for the solution;

generating a deployment sequence of the middleware components;

creating the virtual machines with the initial configuration;

installing a progress monitor on each virtual machine;

initiating deployment of the middleware components to the virtual machines;

monitoring the progress of installation of each middleware component on a virtual machine during said deployment;

computing a deviation of a predictive value associated with said selected configuration policy from an actual value based on said monitoring;

recomputing an objective function of said selected configuration policy;

dynamically adjusting the initial configuration of said virtual machines during said deployment to satisfy said selected configuration policy based on said recomputing; and

installing the solution on the middleware components.

6. A method according to claim 5, wherein said selected configuration policy comprises maximization of server utilization for servers on which the virtual machines are setup.

7. A method according to claim 5, wherein said selected configuration policy comprises provisioning of software components requiring external access based on availability of an external server.

8. A method according to claim 5, wherein said selected configuration policy comprises completion of a base install first and components that can be shared by applications later.

9. A method as recited in claim 5, wherein dynamically adjusting the initial configuration of said virtual machines comprises adjusting central processing unit shares of each virtual machine.

10. A method as recited in claim 5, wherein dynamically adjusting the initial configuration of said virtual machines comprises adjusting a memory size of each virtual machine.

11. A method as recited in claim 5, wherein dynamically adjusting the initial configuration of said virtual machines comprises adjusting a disk size of each virtual machine.

12. A method according to claim 5, wherein the selected configuration policy comprises minimization of total time required to install the middleware components.

13. A method according to claim 5, comprising:

computing an actual time to install the middleware components on the virtual machines;

receiving events or monitoring logs from different progress monitors;

computing a deviation of a predicted time from the actual time;

recomputing the configuration of said virtual machines; and

dynamically adjusting the configuration of said virtual machines to satisfy said selected configuration policy.

14. A method according to claim 5, wherein said dynamically adjusting comprises tuning central processing unit shares, memory size, network access speed, or hard disk access speed for the virtual machines.

15. A method according to claim 5, comprising:

receiving an application solution request;

retrieving a dependency graph listing the middleware components for the application solution;

computing an actual time to install the middleware components on the virtual machines via a table comprising information about times to install software products; and

determining the deployment sequence of the middleware components and the initial configuration of the virtual machines via a table comprising algorithms for different configuration policies.