Co-allocating Compute Resources by Completion Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current resource management in compute environments, such as grids and clusters, faces challenges in optimizing the allocation of different resource types based on completion time and relative machine speed, leading to suboptimal utilization and scheduling of resources.
Innovation Solution
A system and method that analyze resource requests to determine completion times and relative resource speeds, allowing for the co-allocation of resources by calculating a co-allocation parameter that takes into account the speed of execution and completion time, rather than just availability, to reserve resources effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If resources are reserved based on availability only, then resource allocation is simple, but resource utilization is suboptimal
Solution Approach 1:
The patent changes the reservation parameter from simple availability to a composite parameter that includes completion time and relative machine speed. This allows the system to optimize resource utilization by selecting resources that will complete jobs fastest, rather than just any available resource, thereby resolving the contradiction between simple allocation and optimal utilization.
Solution Approach 2:
The system performs preliminary analysis of completion times and relative speeds before making resource reservations. By pre-calculating these parameters and storing them in data structures, the system prepares optimization data in advance, enabling better resource utilization decisions without adding significant complexity to the actual allocation process.
2Productivity
If resources are allocated without considering completion time, then allocation is faster, but job completion efficiency decreases
Solution Approach 1:
The patent calculates and stores completion times and relative speeds in advance in data structures before allocation decisions are made. This preliminary computation allows the system to make optimized allocation decisions quickly without performing complex calculations at allocation time, thus improving job completion efficiency without significant loss of allocation time.
3Productivity
If different resource types are managed separately, then management is simpler, but co-allocation efficiency is reduced
Solution Approach 1:
The patent merges the management of different resource types by introducing a co-allocation parameter that considers completion times and relative speeds across multiple resource types together. Instead of managing each resource type separately, the system integrates them into a unified allocation framework that optimizes the combination of resources, thereby improving co-allocation efficiency while managing complexity through a structured approach.
Data Source
AI summary
A system and method of reserving resources in a compute environment are disclosed. The method embodiment comprises receiving a request for resources within a computer environment, determining at least one completion time associated with at least one resource type required by the request, and reserving resources within the computer environment based on the determine of at least the completion time. A scaled wall clock time on a per resource basis may also be used to determine what resources to reserve. The system may determine whether to perform a start time analysis or a completion time analysis or a hybrid analysis in the process of generating a co-allocation map between a first type of resource and a second type of resource in preparation for reserving resources according to the generated co-allocation map.