Resource Manager Compatibility Attributes Heterogeneous Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current heterogeneous computing environments are limited in resource allocation due to the strong affinity between grid nodes and job execution environments, restricting the use of resources to only those with the same architecture and operating system, leading to underutilization of available nodes.
Innovation Solution
A method where a resource manager in the computing environment obtains compatibility attributes from nodes, including supported platforms and operating systems, to broaden the scope of nodes that can process requests, allowing nodes with different native environments to be considered for resource allocation, using techniques like APIs and Application Binary Interface (ABI) for compatibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If nodes are selected based on matching environment characteristics (same architecture and operating system), then job execution reliability is improved, but resource allocation flexibility deteriorates
Solution Approach 1:
The patent introduces an environment translation layer that acts as an intermediary between the job's required environment and the node's native environment. This translation layer enables jobs compiled for one environment to execute on nodes with different architectures or operating systems, thus maintaining execution reliability while expanding resource allocation flexibility to include heterogeneous nodes.
Solution Approach 2:
The patent makes computing nodes universal by enabling them to execute jobs from multiple environments. Through the environment translation mechanism, a single node can serve multiple job types that would traditionally require separate homogeneous nodes, thereby improving resource allocation flexibility without sacrificing the ability to execute jobs reliably.
2Reliability
If resource allocation is restricted to nodes with matching environment characteristics, then execution compatibility is improved, but resource utilization deteriorates
Solution Approach 1:
The environment translation layer serves as a mediator that allows incompatible jobs to run on heterogeneous nodes. By translating the execution environment, the system maintains execution compatibility while enabling the use of previously unusable nodes, thus improving resource utilization without compromising job execution.
Solution Approach 2:
The patent changes the execution parameters of jobs by translating them from their native environment parameters to the target node's environment parameters. This parameter transformation enables jobs to execute on nodes with different architectures or operating systems, thereby improving resource utilization while maintaining execution compatibility through the translation process.
Data Source
AI summary
Allocation of resources in a heterogeneous computing environment is facilitated. A resource manager of the heterogeneous computing environment obtains information that describes which nodes of the heterogeneous computing environment are capable of supporting additional operating systems and platforms over and above their native operating systems and platforms. This information is then used by the resource manager in allocating resources to satisfy a particular job.