Efficient execution of jobs in a shared pool of resources
US-8972983-B2 · Mar 3, 2015 · US
US9755987B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9755987-B2 |
| Application number | US-201414173503-A |
| Country | US |
| Kind code | B2 |
| Filing date | Feb 5, 2014 |
| Priority date | Feb 5, 2014 |
| Publication date | Sep 5, 2017 |
| Grant date | Sep 5, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A task-oriented virtual resource can be dynamically mapped to multiple physical resource types in order to better align resource allocation mechanisms. Mappings between the task-oriented virtual resource and the physical resource types may be managed in accordance with a workload associated with an underlying task being performed by the network on behalf of a user. The number of physical resource units that are mapped to the task-oriented virtual resource may be varied in accordance with one or more workload conditions using a finite state machine. The finite state machine can include nodes representing static or dynamic states for each physical resource type. The nodes may be interconnected via relationships, which can be governed by the workload conditions. Relationships between nodes may be modeled using hybrid graphs.
Opening claim text (preview).
I claim: 1. A method for sharing resources in a network, the method comprising: creating, by an abstraction layer, a single virtual resource corresponding to a task to be performed by the network on behalf of a customer; mapping, by a network device, the single virtual resource to two or more different types of physical resources, in a resource pool, required to perform the task, the single virtual resource being mapped to a respective number of physical resource units for each of the different types of physical resources; detecting a variation in a workload associated with the task; and dynamically updating numbers of physical resource units for the different types of physical resources mapped to the single virtual resource in accordance with the variation in the workload associated with the task. 2. The method of claim 1 , wherein the single virtual resource is defined by the task to be performed by the network on behalf of the customer. 3. The method of claim 1 , wherein physical resources in the resource pool are classified as having one of a plurality of resource types, the plurality of resource types including at least a processing resource type and a memory resource type, and wherein the single virtual resource comprises a task-oriented resource type that is excluded from the plurality of resource types used to classify physical resources in the resource pool. 4. The method of claim 1 , wherein mapping the single virtual resource to multiple physical resources in the resource pool of the network comprises: mapping the single virtual resource to a number of physical processing resource units and a number of physical memory resource units in accordance with an initial value of the workload associated with the task. 5. The method of claim 4 , wherein the number of physical processing resource units mapped to the single virtual resource is different than the number of physical memory resource units mapped to the single virtual resource. 6. The method of claim 4 , wherein dynamically updating the number of physical resource units mapped to the single virtual resource in accordance with the variation in the workload associated with the task comprises: dynamically updating the number of physical processing resource units mapped to the single virtual resource or the number of physical memory resource units mapped to the single virtual resource in accordance with the variation in the workload associated with the task. 7. The method of claim 1 , wherein mapping the single virtual resource to multiple physical resources in the resource pool comprises: mapping the single virtual resource to physical resources in accordance with a hybrid graph, the hybrid graph modeling relationships between workloads associated with the task, the single virtual resource, and the multiple physical resources. 8. The method of claim 1 , further comprising creating a finite state machine for managing the single virtual resource. 9. A method for sharing resources in a network, the method comprising: creating, by an abstraction layer, a single virtual resource, the single virtual resource corresponding to a task to be performed by the network on behalf of a customer; mapping, by a network device, the single virtual resource to multiple physical resources in a resource pool of the network; and creating a finite state machine for managing the single virtual resource, wherein creating the finite state machine for managing the single virtual resource comprises: identifying, by the abstraction layer, a physical resource type classifying a number of physical resource units being mapped to the single virtual resource; creating, by the abstraction layer, a static state and a dynamic state for the physical resource type, wherein the number of physical resource units being mapped to the single virtual resource is held constant when the finite state machine operates in the static state, and wherein the number of physical resource units being mapped to the single virtual resource is varied when the finite state machine operates in the dynamic state; and creating, by the abstraction layer, a relationship between the static state and the dynamic state in accordance with a workload assigned to the task. 10. The method of claim 9 , wherein the relationship defines a triggering condition for migrating the finite state machine from the static state to the dynamic state or vice versa. 11. The method of claim 10 , further comprising: identifying, by the abstraction layer, a first physical resource type classifying a first type of physical resource units being mapped to the single virtual resource; identifying, by the abstraction layer, a second physical resource type classifying a second type of physical resource units being mapped to the single virtual resource; creating, by the abstraction layer, a first dynamic state for the first physical resource type, wherein a number of first type physical resource units being mapped to the single virtual resource is varied when the finite state machine operates in the first dynamic state; creating, by the abstraction layer, a second dynamic state for the second physical resource type, wherein a number of second type physical resource units being mapped to the single virtual resource is varied when the finite state machine operates in the second dynamic state; and creating, by the abstraction layer, a relationship between the first dynamic state and the second dynamic state in accordance with a workload parameter of the task associated with the single virtual resource. 12. The method of claim 11 , wherein the relationship defines a triggering condition for migrating the finite state machine from the first dynamic state to the second dynamic state. 13. A non-transitory computer program product comprising a computer readable storage medium storing programming, the programming including instructions to: create a single virtual resource corresponding to a task to be performed by a network on behalf of a customer; map the single virtual resource to two or more different types of physical resources, in a resource pool, required to perform the task, the single virtual resource being mapped to a respective number of physical resource units for each of the different types of physical resources; detect a variation in a workload associated with the task; and dynamically update numbers of physical resource units for the different types of physical resources mapped to the single virtual resource in accordance with the variation in the workload associated with the task. 14. The non-transitory computer program product of claim 13 , wherein the instructions to map the single virtual resource to multiple physical resources in the resource pool includes instructions to: map the single virtual resource to physical resources in accordance with a hybrid graph, the hybrid graph modeling relationships between workloads assigned to the task, the single virtual resource, and the multiple physical resources. 15. The non-transitory computer program product of claim 13 , wherein physical resources in the resource pool are classified as having one of a plurality of resource types, the plurality of resource types including at least a processing resource type and a memory resource type, and wherein the single virtual resource comprises a task-oriented resource type that is excluded from the plurality of resource types used to classify physical resources in the resource pool. 16. A non-transitory computer program product comprising a computer readable storage medium storing programming, the programming including
Hypervisors; Virtual machine monitors · CPC title
Admission control; Resource allocation · CPC title
Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities (flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76) · CPC title
by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.