Dynamically allocating compute nodes among cloud groups based on priority and policies

Patent metadata

Field	Value
Publication number	US-9571374-B2
Application number	US-201514594295-A
Country	US
Kind code	B2
Filing date	Jan 12, 2015
Priority date	Jan 12, 2015
Publication date	Feb 14, 2017
Grant date	Feb 14, 2017

Abstract

A method, system and computer program product for allocating compute nodes among cloud groups. Policies are created for cloud groups for specifying conditions when a compute node will need to be borrowed by a cloud group as well as when the borrowed compute node is to be returned. Furthermore, priorities are assigned to the conditions in the policies for borrowing a compute node as well as to the cloud groups concerning borrowing compute nodes. A compute node is allocated temporarily from a first cloud group to a second cloud group based on the priority assigned to the second cloud group and the priority assigned to the monitored condition invoking the borrowing of the compute node as well as based on the monitored condition satisfying a condition in the policy for the second cloud group as to when the compute node will need to be borrowed by the second cloud group.

First claim

The invention claimed is:

1. A method for allocating compute nodes among cloud groups, the method comprising:
creating policies for task-specific cloud groups for specifying conditions when a compute node will need to be borrowed by a task-specific cloud group as well as when said borrowed compute node is to be returned, wherein said borrowed compute node is a compute node assigned as a backup resource for one or more task-specific cloud groups, wherein each of said task-specific cloud groups comprises a plurality of compute nodes assigned to host a designated workload type;
assigning priorities to said conditions in said policies for borrowing said compute node as well as to said task-specific cloud groups concerning borrowing said compute node;
monitoring conditions of a cloud computing environment;
allocating, by a processor, said borrowed compute node from a first task-specific cloud group to a second specific cloud group based on said priority assigned to said second task-specific cloud group, said priority assigned to a monitored condition of said second task-specific cloud group invoking said borrowing of said compute node and said monitored condition of said second task-specific cloud group satisfying a condition in said policy for said second task-specific cloud group as to when said borrowed compute node will need to be borrowed by said second task-specific cloud group; and
migrating a workload to said borrowed compute node.

2. The method as recited in claim 1 further comprising:
reviewing said policies to determined when said borrowed compute node is to be returned to said first task-specific cloud group.

3. The method as recited in claim 1 , wherein said task-specific cloud groups host workload types of production, development and testing.

4. The method as recited in claim 1 , wherein said monitored condition comprises one of the following: a hardware failure, an expected hardware failure, a scaling requirement, a capacity shortage, a spare capacity to apply maintenance, a workload to swap out, an absence of a resource and a peak utilization.

5. The method as recited in claim 1 , wherein said borrowed compute node is a virtual compute node.

6. The method as recited in claim 1 , wherein said borrowed compute node is a physical compute node.

7. A computer program product for allocating compute nodes among cloud groups, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code comprising the programming instructions for:
creating policies for task-specific cloud groups for specifying conditions when a compute node will need to be borrowed by a task-specific cloud group as well as when said borrowed compute node is to be returned, wherein said borrowed compute node is a compute node assigned as a backup resource for one or more task-specific cloud groups, wherein each of said task-specific cloud groups comprises a plurality of compute nodes assigned to host a designated workload type;
assigning priorities to said conditions in said policies for borrowing said compute node as well as to said task-specific cloud groups concerning borrowing said compute node;
monitoring conditions of a cloud computing environment;
allocating said borrowed compute node from a first task-specific cloud group to a second specific cloud group based on said priority assigned to said second task-specific cloud group, said priority assigned to a monitored condition of said second task-specific cloud group invoking said borrowing of said compute node and said monitored condition of said second task-specific cloud group satisfying a condition in said policy for said second task-specific cloud group as to when said borrowed compute node will need to be borrowed by said second task-specific cloud group; and
migrating a workload to said borrowed compute node.

8. The computer program product as recited in claim 7 , wherein the program code further comprises the programming instructions for:
reviewing said policies to determined when said borrowed compute node is to be returned to said first task-specific cloud group.

9. The computer program product as recited in claim 7 , wherein said task-specific cloud groups host workload types of production, development and testing.

10. The computer program product as recited in claim 7 , wherein said monitored condition comprises one of the following: a hardware failure, an expected hardware failure, a scaling requirement, a capacity shortage, a spare capacity to apply maintenance, a workload to swap out, an absence of a resource and a peak utilization.

11. The computer program product as recited in claim 7 , wherein said borrowed compute node is a virtual compute node.

12. The computer program product as recited in claim 7 , wherein said borrowed compute node is a physical compute node.

13. A system, comprising:
a memory unit for storing a computer program for allocating compute nodes among cloud groups; and
a processor coupled to the memory unit, wherein the processor is configured to execute the program instructions of the computer program comprising:
creating policies for task-specific cloud groups for specifying conditions when a compute node will need to be borrowed by a task-specific cloud group as well as when said borrowed compute node is to be returned, wherein said borrowed compute node is a compute node assigned as a backup resource for one or more task-specific cloud groups, wherein each of said task-specific cloud groups comprises a plurality of compute nodes assigned to host a designated workload type;
assigning priorities to said conditions in said policies for borrowing said compute node as well as to said task-specific cloud groups concerning borrowing said compute node;
monitoring conditions of a cloud computing environment;
allocating said borrowed compute node from a first task-specific cloud group to a second specific cloud group based on said priority assigned to said second task-specific cloud group, said priority assigned to a monitored condition of said second task-specific cloud group invoking said borrowing of said compute node and said monitored condition of said second task-specific cloud group satisfying a condition in said policy for said second task-specific cloud group as to when said borrowed compute node will need to be borrowed by said second task-specific cloud group; and
migrating a workload to said borrowed compute node.

14. The system as recited in claim 13 , wherein the program instructions of the computer program further comprises:
reviewing said policies to determined when said borrowed compute node is to be returned to said first task-specific cloud group.

15. The system as recited in claim 13 , wherein said task-specific cloud groups host workload types of production, development and testing.

16. The system as recited in claim 13 , wherein said monitored condition comprises one of the following: a hardware failure, an expected hardware failure, a scaling requirement, a capacity shortage, a spare capacity to apply maintenance, a workload to swap out, an absence of a resource and a peak utilization.

17. The system as recited in claim 13 , wherein said borrowed compute node is a virtual compute node.

18. The system as recited in claim 13 , wherein said borrowed compute node is a physical compute node.

Assignees

• IBM

Inventors

• Ashok Rohith K
• Brabson Roy F
• Heninger Ivan M
• Hockett Hugh E
• Olszewski Bret R
• Quirk Aaron J

Classifications

• H04L41/0893

  Assignment of logical groups to network elements · CPC title

• H04L43/10 Primary

  Active monitoring, e.g. heartbeat, ping or trace-route · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L41/0897

  by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title

• H04L41/0894

  Policy-based network configuration management · CPC title