Network resource assignment system and method

• US 7,310,673 B2
• Filed: 12/21/2001
• Issued: 12/18/2007
• Est. Priority Date: 12/21/2001
• Status: Expired due to Fees

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Zhu, Xiaoyun, Singhal, Sharad
• Primary Examiner(s): Vaughn; William
• Assistant Examiner(s): MESFIN, YEMANE
• Application Number:
  US10/032,105
• Publication Number:
  US 20030120780A1
• Time in Patent Office: 2,188 Days
• Field of Search: 709/202, 709/220, 709/223, 709/232, 709/226, 709/244, 709/224, 709/225, 703/2, 713/201, 370/468
• US Class Current: 709/226
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

The present invention is a system and method for facilitating efficient assignment of networked resources. In one embodiment, a resource assignment method is utilized to assist network resource assignment operations. The resource assignment method establishes a resource model, acquires an application model, and utilizes a mapping process to map said application model onto said resource model. In one exemplary implementation the resources are assigned to optimize the assignment of resources with respect to application requirements and desired objectives (e.g., minimization of the average communication delay inside a service core). In one exemplary implementation, a mapping process of a resource assignment method is utilized to map an application onto part of IDC topology.

Claims

1. A resource assignment method comprising:

establishing a resource model;

acquiring an application model, wherein said application model describes a plurality of application functional components and includes information about which application functional components communicate with each other;

obtaining a set of parameters associated with topology and performance characteristics of resources in a data center, said parameters comprising;

a number of edge switches, a number of rack switches, a number of server nodes, and connectivity matrices between different layers; and

a specification of the bandwidth limits of the incoming and outgoing links at various layers of the network;

acquiring information about resource requirements of an application; and

utilizing a mapping process to map said application model onto said resource model, wherein said mapping process is directed to increasing the optimization of resource utilization through appropriate assignment of resources to an application by identifying infeasible networked resources and partitioning the available networked resources with respect to desired objectives and wherein said mapping process is enabled to map a first application model onto a first resource model and a second application model onto a second resource model.

2. The resource assignment method of claim 1 wherein said information about resource requirements of an application include:

a number of application functional components;

a network traffic requirements between said application functional components; and

upper and lower bounds on server attributes which are required for said server to host said application functional component.

3. The resource assignment method of claim 1 wherein said mapping process determines which server nodes are assigned to an application functional component and is captured in an assignment decision variable.

4. The resource assignment method of claim 3 wherein said assignment decision variable is optimized in accordance with a desired objective including meeting application requirements.

5. The resource assignment method of claim 3 wherein said desired objective further includes minimizing communication delays.

6. The resource assignment method of claim 3 wherein a layered partitioning and pruning (LPP) process is utilized to find an application resource assignment optimal solution.

7. A resource allocation system comprising:

a means for gathering information parameters associated with topology and performance characteristics of available networked resources, said parameters comprising;

a number of edge switches, a number of rack switches, a number of server nodes, and connectivity matrices between different layers; and

a specification of the bandwidth limits of the incoming and outgoing links at various layers of the network;

a means for extracting information associated with application functional components and resource requirements; and

a means for assigning application functional components to said available networked resources including a means identifying infeasible networked resources and partitioning the available networked resources in accordance with a resource allocation variable, wherein said means for assigning uses information about which application functional components communicate with each other as a part of assigning said application functional components to said available networked resources and wherein said means for assigning enables assigning first information associated with a first set of available networked resources with first information associated with a first set of application functional components and assigning second information associated with a second set of available networked resources with second information associated with a second set of application functional components.

8. A resource allocation system of claim 7 wherein said means for assigning application functional components to said available networked resources allocates, said available networked resources by maximizing said available networked resources identified in said resource allocation variable with respect to application constraints and desired objectives.

9. The resource allocation system of claim 7 wherein said information associated with said available networked resources includes configuration and performance characteristics of said available networked resources.

10. The resource allocation system of claim 9 wherein said information associated with said application functional components includes organization and networked resource requirements of said application functional components.

11. The resource allocation system of claim 10 wherein said means for assigning application functional components to said available networked resources includes a means for simplifying said assignment analysis by identifying infeasible networked resources and partitioning said available networked resources.