Cloud Workload Redeployment Through Elastic Computing Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional single-machine computing is inadequate for drug research and development, requiring high computing power with variable demands, leading to inefficiency and high fixed costs when idle resources are common.
Innovation Solution
A cloud-based data processing system with elastic scaling capabilities, allowing dynamic allocation and redeployment of computing resources based on demand, using a distributed task scheduling system and cloud elastic scaling system to manage workload and computing nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If a large computer room is set up forscientific computing, then computing power requirement is met, but fixed cost is too high and computing resources are wasted when idle
Solution Approach 1:
The patent implements dynamic scaling of computing resources based on actual workload demands. The system automatically adjusts the number of computing nodes and resource allocation in real-time, transitioning from static fixed-cost infrastructure to dynamic on-demand resource provisioning. This resolves the contradiction by enabling high computing power when needed while eliminating resource wastage during idle periods through automatic resource termination.
Solution Approach 2:
The system changes the parameter of resource allocation from fixed to variable based on workload characteristics. By monitoring task submission patterns and computing resource utilization, the system dynamically adjusts infrastructure parameters (number of nodes, resource types, configuration) to match actual demands, thereby meeting computing power requirements when needed while avoiding fixed-cost waste during low-demand periods.
2Quantity of substance
If computing resources are fixed, then infrastructure cost is controlled, but computing requirements are very variable causing great waste when idle
Solution Approach 1:
The patent transforms fixed computing resources into dynamic, on-demand resources through automated scaling mechanisms. The system continuously monitors workload patterns and automatically provisions or terminates computing nodes based on real-time demands, enabling the infrastructure to adapt to variable computing requirements without manual intervention or fixed commitments.
Solution Approach 2:
The system implements feedback loops that monitor workload submission patterns, resource utilization metrics, and task completion status. This feedback information drives automated decisions about resource allocation and scaling, allowing the system to respond adaptively to changing computational demands while optimizing resource utilization and eliminating waste during idle periods.
3Loss of energy
If single-machine computing is used, then cost is low, but computing power is insufficient and tasks take weeks to months
Solution Approach 1:
The patent merges multiple computing nodes into a coordinated distributed computing cluster that operates as a unified high-performance system. By combining the computing power of multiple machines working in parallel on the same scientific computations, the system achieves the computing speed of a supercomputer while maintaining the cost-effectiveness of standard hardware components and eliminating the need for expensive dedicated supercomputer infrastructure.
Data Source
AI summary
The present application provides a data processing method, system, electronic equipment and storage medium based on a cloud platform, which are applied to the technical field of cloud computing processing, wherein the data processing method comprises the following steps: obtaining task processing requests submitted by several target users through a distributed system, wherein the task processing requests are requests for processing scientific computing tasks; Determining whether the number of the obtained task processing requests reaches a preset capacity expansion threshold, and if so, generating a workload capacity expansion request; Performing capacity expansion processing on the computing node according to the capacity expansion request; The workload is redeployed based on the computing nodes after capacity expansion processing, so as to execute the scientific computing task based on the redeployed workload.