Framework for QoS in embedded computer infrastructure
First Claim
Patent Images
1. A method providing a framework for Quality of Service (QoS) within and between globally distributed computing components by a processor, the method comprising:
- identifying a QoS for a particular user by comparing identification information with information in a predefined table;
defining a resource template for the user based on the identified QoS, the resource template specifying computing resources required for a computing process uploaded by the user to an object store;
appending information, including header and metadata information associated with the computing process, to the resource template;
estimating at least one resource required for the computing process by examining information associated with the resource template and executing the computing process from within a sandboxed environment in the object store to gauge resource utilization prior to accepting the computing process for general deployment, wherein the computing process is accepted for general deployment when the resource utilization aligns with constraints delineated by the resource template, andupon accepting the computing process for general deployment;
allocating a storlet as the at least one resource at a storage node, the storlet comprising embedded compute functionality within the object store and thereby offloading computing elements to at least one storage unit;
wherein the allocated storlet performs the computing process according to the constraints delineated by the resource template to efficiently manage and bound the computing resources between multiple tenants of the object store.
1 Assignment
0 Petitions
Accused Products
Abstract
Various embodiments providing a framework for Quality of Service (QoS) within and between globally distributed computing components by a processor. At least one resource required for a computing process is estimated by examining information associated with a resource template. A storlet is allocated as the at least one resource at a storage node, thereby offloading computing elements to at least one storage unit. The allocated storlet performs the computing process according to constraints delineated by the resource template.
24 Citations
18 Claims
-
1. A method providing a framework for Quality of Service (QoS) within and between globally distributed computing components by a processor, the method comprising:
-
identifying a QoS for a particular user by comparing identification information with information in a predefined table; defining a resource template for the user based on the identified QoS, the resource template specifying computing resources required for a computing process uploaded by the user to an object store; appending information, including header and metadata information associated with the computing process, to the resource template; estimating at least one resource required for the computing process by examining information associated with the resource template and executing the computing process from within a sandboxed environment in the object store to gauge resource utilization prior to accepting the computing process for general deployment, wherein the computing process is accepted for general deployment when the resource utilization aligns with constraints delineated by the resource template, and upon accepting the computing process for general deployment; allocating a storlet as the at least one resource at a storage node, the storlet comprising embedded compute functionality within the object store and thereby offloading computing elements to at least one storage unit; wherein the allocated storlet performs the computing process according to the constraints delineated by the resource template to efficiently manage and bound the computing resources between multiple tenants of the object store. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A system providing a framework for Quality of Service (QoS) within and between globally distributed computing components, comprising:
-
a processor, coupled to a memory which stores computer readable program instructions that when executed by the processor; identifies a QoS for a particular user by comparing identification information with information in a predefined table, defines a resource template for the user based on the identified QoS, the resource template specifying computing resources required for a computing process uploaded by the user to an object store, appends information, including header and metadata information associated with the computing process, to the resource template, estimates at least one resource required for Hall the computing process by examining information associated with Hall the resource template and executing the computing process from within a sandboxed environment in the object store to gauge resource utilization prior to accepting the computing process for general deployment, wherein the computing process is accepted for general deployment when the resource utilization aligns with constraints delineated by the resource template, and upon accepting the computing process for general deployment; allocates a storlet as the at least one resource at a storage node, the storlet comprising embedded compute functionality within the object store and thereby offloading computing elements to at least one storage unit; wherein the allocated storlet performs the computing process according to the constraints delineated by the resource template to efficiently manage and bound the computing resources between multiple tenants of the object store. - View Dependent Claims (8, 9, 10, 11, 12)
-
-
13. A computer program product providing a framework for Quality of Service (QoS) within and between globally distributed computing components by a processor, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
-
an executable portion that identifies a QoS for a particular user by comparing identification information with information in a predefined table; an executable portion that defines a resource template for the user based on the identified QoS, the resource template specifying computing resources required for a computing process uploaded by the user to an object store; an executable portion that appends information, including header and metadata information associated with the computing process, to the resource template; an executable portion that estimates at least one resource required for the computing process by examining information associated with the resource template and executing the computing process from within a sandboxed environment in the object store to gauge resource utilization prior to accepting the computing process for general deployment, wherein the computing process is accepted for general deployment when the resource utilization aligns with constraints delineated by the resource template, and upon accepting the computing process for general deployment; an executable portion that allocates a storlet as the at least one resource at a storage node, the storlet comprising embedded compute functionality within the object store and thereby offloading computing elements to at least one storage unit; wherein the allocated storlet performs the computing process according to the constraints delineated by the resource template to efficiently manage and bound the computing resources between multiple tenants of the object store. - View Dependent Claims (14, 15, 16, 17, 18)
-
Specification