Dynamic Cloud Resource Allocation for Heterogeneous Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In heterogeneous cloud computing environments, batch computing resources are often underutilized due to the difficulty in adapting to diverse computing resources, leading to low efficiency in processing tasks, as existing methods require complex user intervention or limited support for network-intensive jobs.
Innovation Solution
A scheduling method and system that automatically allocates actual computing resources in excess of the initial allocation based on a resource description manifest, allowing for dynamic utilization of available resources without manual job splitting, thereby improving resource utilization and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If single resource specifications are used to allocate computing resources in heterogeneous cloud computing environments, then resource allocation is simple, but computing resources are underutilized
Solution Approach 1:
The patent implements dynamic resource allocation where the system automatically adjusts resource assignment based on actual hardware configurations. The batch computing system detects heterogeneous resource characteristics and dynamically assigns tasks to match available CPU cores, memory, and storage capacities, transforming static single-specification allocation into adaptive multi-specification allocation that resolves the contradiction between simplicity and utilization efficiency
Solution Approach 2:
The system changes allocation parameters by introducing multi-dimensional resource specification matching. Instead of using a single resource template, the patent employs multiple resource specification parameters (CPU cores, memory size, storage capacity) that can be dynamically adjusted to match heterogeneous cloud computing resources, thereby improving resource utilization without significantly increasing allocation complexity
2Adaptability or versatility
If users split jobs into multiple different jobs to adapt to heterogeneous resources, then resource adaptation improves, but operation complexity increases
Solution Approach 1:
The patent implements self-service resource adaptation where the batch computing system automatically detects heterogeneous resource configurations and performs job scheduling without requiring user intervention. The system autonomously matches tasks to appropriate resource specifications, eliminating the need for users to manually split jobs and thereby maintaining high adaptability while reducing operational complexity to zero
Solution Approach 2:
The patent introduces an intermediary resource management layer between the user's batch computing jobs and the heterogeneous cloud resources. This intermediary automatically handles the adaptation process by translating user job requirements into resource-specific execution parameters, serving as a mediator that resolves the contradiction by absorbing the complexity of resource adaptation away from the user
3Power
If higher configuration resources are used for each instance, then computing power per instance increases, but the number of usable computing nodes decreases
Solution Approach 1:
The patent applies local quality by assigning different resource configurations to different instances based on their specific requirements and the available heterogeneous resources. Instead of uniformly allocating high-configuration resources to all instances, the system matches each instance's computing needs with appropriate local resource characteristics, thereby maximizing both the computing power utilized per instance and the total number of nodes that can be effectively employed
Data Source
AI summary
A device and method for automatically allocating computing resources is disclosed herein. The method includes receiving a task from a client, the task including a plurality of instances and a resource description manifest representing resource needs of the plurality of instances; determining an initial computing resource allocation of a cluster of machines based on the resource description manifest, wherein the initial computing resource allocation is determined based on the resource needs included in the resource description manifest; determining that the resource description manifest indicates a request to utilize an actual computing resource allocation in excess of the initial computing resource allocation; configuring a plurality of actual computing resources to process the plurality of instances, wherein the plurality of actual computing resources are configured to utilize resources in excess of the initial computing resource allocation; and executing the plurality of instances using the plurality of actual computing resources.