Dynamically composed compute nodes comprising disaggregated components

Patent metadata

Field	Value
Publication number	US-10382279-B2
Application number	US-201414320069-A
Country	US
Kind code	B2
Filing date	Jun 30, 2014
Priority date	Jun 30, 2014
Publication date	Aug 13, 2019
Grant date	Aug 13, 2019

Abstract

A method, article of manufacture, and apparatus for creating dynamically composed compute nodes from disaggregated hardware components is discussed. These components may be dynamically allocated from resource pools to the compute nodes.

First claim

What is claimed is:

1. A method for managing resources in a disaggregated infrastructure, the method comprising:
pooling disaggregated hardware resources across multiple enclosures into resource pools, the resource pools comprising a compute resource pool, a memory resource pool, and a storage resource pool, wherein each of the multiple enclosures includes a compute resource in the compute pool, a memory resource in the memory pool, and a storage resource in the storage pool, wherein the disaggregated hardware resources are in communication with a compute fabric, wherein the disaggregated hardware resources communicate directly with each other using the compute fabric when they are in different enclosures within the multiple enclosures, wherein a first subset of the resource pools are reserved for specific users or processes and a second subset of the resource pools comprise general purpose resource pools;
receiving a first workload request at a management system associated with the disaggregated infrastructure;
determining by the management system a first resource requirement from the first workload request, wherein the first resource requirement comprises a power requirement;
allocating dedicated resources to a dynamically composed computed node (“DCCN”) from the resource pools, wherein the allocation is based on the first resource requirement;
receiving a second workload request at the management system associated with the disaggregated infrastructure;
determining a second resource requirement from the second workload request, wherein the second resource requirement comprises a power requirement;
allocating or de-allocating dedicated resources to or from the DCCN from or to the resource pools, wherein the allocating or deallocating is based on the second resource requirement; and
dynamically allocating or deallocating dedicated resources for the DCCN from or to the resource pools based on the management system monitoring the DCCN, wherein the management system has visibility into the resource pools allowing the management system to provide information about the disaggregated hardware resources and their utilization, wherein the management utility may dynamically switch between in-band monitoring and/or out-of-band monitoring.

2. The method of claim 1 , wherein the disaggregated infrastructure comprises a plurality of hardware racks comprising hardware components, wherein the hardware components are in communication with each other through the compute fabric.

3. The method of claim 1 , wherein the compute resource pool comprises compute resources, the memory resource pool comprises memory resources, and the storage resource pool comprises storage resources.

4. The method of claim 1 , further comprising dividing the disaggregated infrastructure into a plurality of views, wherein individual views describe components encapsulated by the view.

5. The method of claim 4 , further comprising nesting a first view inside a second view, wherein the first view describes components encapsulated by both the first view and the second view.

6. A computer program product for managing resources in a disaggregated infrastructure, the computer program product comprising a non-transitory computer readable medium encoded with computer executable program, the code enabling:
pooling disaggregated hardware resources across multiple enclosures into resource pools, the resource pools comprising a compute resource pool, a memory resource pool, and a storage resource pool, wherein each of the multiple enclosures includes a compute resource in the compute pool, a memory resource in the memory pool, and a storage resource in the storage pool, wherein the disaggregated hardware resources are in communication with a compute fabric, wherein the disaggregated hardware resources communicate directly with each other using the compute fabric when they are in different enclosures within the multiple enclosures, wherein a first subset of the resource pools are reserved for specific users or processes and a second subset of the resource pools comprise general purpose resource pools;
receiving a first workload request at a management system associated with the disaggregated infrastructure;
determining a first resource requirement from the first workload request, wherein the first resource requirement comprises a power requirement;
allocating dedicated resources to a dynamically composed computed node (“DCCN”) from the resource pools, wherein the allocation is based on the first resource requirement;
receiving a second workload request at the management system associated with the disaggregated infrastructure;
determining a second resource requirement from the second workload request, wherein the second resource requirement comprises a power requirement;
allocating or de-allocating dedicated resources to or from the DCCN from or to the resource pools, wherein the allocating or deallocating is based on the second resource requirement; and
dynamically allocating or deallocating dedicated resources for the DCCN from or to the resource pools based on the management system monitoring the DCCN, wherein the management system has visibility into the resource pools allowing the management system to provide information about the disaggregated hardware resources and their utilization, wherein the management utility may dynamically switch between in-band monitoring and/or out-of-band monitoring.

7. The computer program product of claim 6 , wherein the disaggregated infrastructure comprises a plurality of hardware racks comprising hardware components, wherein the hardware components in the racks are in communication with each other through the compute fabric.

8. The computer program product of claim 6 , further comprising dividing the disaggregated infrastructure into a plurality of views, wherein individual views describe components encapsulated by the view.

9. The computer program product of claim 8 , further comprising nesting a first view inside a second view, wherein the first view describes components encapsulated by both the first view and the second view.

10. A system for managing resources in a disaggregated infrastructure, the computer system comprising a computer processor configured to execute instructions comprising:
pooling disaggregated hardware resources across multiple enclosures into resource pools, the resource pools comprising a compute resource pool, a memory resource pool, and a storage resource pool, wherein each of the multiple enclosures includes a compute resource in the compute pool, a memory resource in the memory pool, and a storage resource in the storage pool, wherein the disaggregated hardware resources are in communication with a compute fabric, wherein the disaggregated hardware resources communicate directly with each other using the compute fabric when they are in different enclosures within the multiple enclosures, wherein a first subset of the resource pools are reserved for specific users or processes and a second subset of the resource pools comprise general purpose resource pools;
receiving a first workload request at a management system associated with the disaggregated infrastructure;
determining a first resource requirement from the first workload request, wherein the first resource requirement comprises a power requirement;
allocating dedicated resources to a dynamically composed computed node (“DCCN”) from the resource pools, wherein the allocation is based on the first resource requirement;
receiving a second workload request at the management system associated with the disaggregated infrastructure;
determining a second resource requirement from the second workload request, wherein the second resource requirement

Assignees

• Emc Corp

Inventors

• Roese John
• Durazzo Kenneth
• Todd Stephen

Classifications

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• H04L41/0896 Primary

  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