System and method for providing a scalable on demand hosting system

• US 7,437,730 B2
• Filed: 11/14/2003
• Issued: 10/14/2008
• Est. Priority Date: 11/14/2003
• Status: Active Grant

• Current Assignee: Daedalus Blue LLC
• Original Assignee: International Business Machines Corporation
• Inventors: Goyal, Pawan
• Primary Examiner(s): Bullock, Jr.; Lewis A.
• Assistant Examiner(s): To; Jennifer N
• Application Number:
  US10/714,331
• Publication Number:
  US 20050108712A1
• Time in Patent Office: 1,796 Days
• Field of Search: 718/1, 718100-108, 709201-203, 709223-226
• US Class Current: 718/105
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

A VM based hosting architecture system in which finer grain control in optimizing multiple workloads across multiple servers is provided. The system includes a plurality of servers to be utilized by multiple workloads. In addition, the system includes a plurality of virtual machines (VMs) at each of the plurality of servers, wherein the plurality of VMs at each of the plurality of servers each serve a different one of the multiple workloads. Moreover, the system includes resource management logic to distribute server resources to each of the plurality of VMs according to predicted resource needs of each of the multiple workloads. Each of the multiple workloads are distributed across the plurality of servers, wherein fractions of each of the multiple workloads are handled by the plurality of VMs. The distribution of multiple workloads over multiple servers has the effect of achieving a finer grain control in optimizing workloads across the plurality of servers.

Claims

1. A system to provide finer grain control in optimizing multiple workloads across multiple servers, comprising:

a plurality of servers to be utilized by multiple workloads;

a plurality of virtual machines at each of the plurality of servers, wherein the plurality of virtual machines at each of the plurality of servers each serve a different one of the multiple workloads; and

resource management logic to distribute server resources to each of the plurality of virtual machines according to current and predicted resource needs of each of the multiple workloads utilizing the server resources,

whereby, each of the multiple workloads are distributed across the plurality of servers, wherein fractions of each of the multiple workloads are handled by the plurality of virtual machines,

whereby, the fractions of each of the multiple workloads handled by each of the virtual machines can be dynamically adjusted to provide for optimization of the server resources utilized by the multiple workloads across the multiple servers.

2. The system of claim 1 wherein the server resources comprise percentage of CPU, percentage of network bandwidth, disk resources and memory resources.

3. The system of claim 1 wherein the finer grain control is achieved through recognizing when one of the plurality of servers is overloaded and shifting work to another of the plurality of servers which is not overloaded.

4. The system of claim 1 wherein the fractions of the multiple workloads being handled by the plurality of virtual machines can be dynamically adjusted in response to workload changes at the plurality of servers, wherein the dynamic adjustment provides for maintaining an optimum utilization level of the server resources utilized by the multiple workloads distributed across the plurality of servers.

5. The system of claim 4 wherein the optimum utilization level can be configured automatically via server management software or manually by a user with administrative privileges.

6. The system of claim 1 wherein the workloads are each distributed over a subset of the plurality of virtual machines.

7. The system of claim 6 wherein each VM in the subset of the plurality of virtual machines exists at a separate one of the plurality of servers.

8. The system of claim 7 wherein the workload distribution comprises distributing the work according to resources available to each of the virtual machines within the subset.

9. The system of claim 1 further comprises at least one global resource allocator to monitor resource distribution between the plurality of virtual machines.

10. The system of claim 9 further comprises at least one load balancer to measure the current offered load.

11. The system of claim 10 wherein the global resource allocator determines how to distribute the resources between the plurality of virtual machines, according to the measurements received from the at least one load balancer.

12. The system of claim 11 wherein each of the plurality of servers includes a local resource control agent to receive and implement instructions from the global resource allocator describing how the resources are to be distributed between the virtual machines located at each of the plurality of servers.

13. A server optimization device having a processor, the server optimization device comprising:

a load balancer associated with a respective customer workload needing at least one workload server and for providing offered workload messages to a provider of workload servers;

a global resource allocator (GRA) for inclusion in said provider of workload servers, and for receiving said offered workload messages and assigning an optimum matching of combinations of whole integer numbers of workload servers and fractional virtual workload servers that the GRA controls to each of the respective customer workloads according to identified resource requirements.

14. The server optimization device of claim 13 wherein the fractional virtual workload servers are reassigned according to changes in the identified resource requirements of the workload assigned.

15. The server optimization device of claim 13 further comprises a resource control agent associated with each of the whole integer number of workload servers, for creating and assigning fractional virtual workload servers to workloads in response to instructions received from the global resource allocator.

16. The server optimization device of claim 13 wherein the identified resource requirements comprise the percentage of CPU, percentage of network bandwidth, disk resources and memory resources that are needed by a workload.