Cluster Workload Scoring to Contain Antagonistic Propagation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing systems face challenges in managing distributed workloads due to antagonistic workloads that rapidly spread across multiple locations, causing performance degradation or outages by negatively impacting shared infrastructure components.
Innovation Solution
A method and system that utilize workload keys based on characteristics to associate workloads with similar execution histories, determining scores for each cluster, and executing workloads at optimal clusters to mitigate antagonistic behavior, with policies to control propagation and isolate potentially harmful workloads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If workloads are rapidly distributed to execute at multiple different locations, then workload execution speed and scalability are improved, but the risk of antagonistic workloads spreading and causing performance degradation increases
Solution Approach 1:
The system performs preliminary actions by generating workload keys and obtaining workload histories before distributing workloads to clusters. This allows the system to pre-assess potential antagonistic behavior and make informed decisions about workload placement, preventing harmful workloads from spreading across multiple locations before damage occurs.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring workload execution results and updating workload histories. This feedback loop allows the system to learn from past workload behaviors and adjust future workload distribution decisions, improving reliability while maintaining execution speed through intelligent, data-driven placement decisions.
2Measurement precision
If workload characteristics are analyzed in detail to identify antagonistic patterns, then detection accuracy is improved, but processing time and computational complexity increase
Solution Approach 1:
The system segments workload analysis by extracting specific key characteristics (workload name, username, source code) and using these segmented features to generate workload keys. This segmentation allows for efficient comparison and matching without requiring complete analysis of entire workload configurations, maintaining detection accuracy while reducing processing time.
Solution Approach 2:
The system changes parameters by transforming detailed workload characteristics into condensed workload keys and scores. This parameter transformation enables rapid comparison and identification of potentially antagonistic workloads without processing all original workload details, balancing detection precision with computational efficiency.
Data Source
AI summary
A method includes receiving a request to execute a particular workload of a plurality of workloads at a distributed computing system that includes a plurality of clusters. Each workload of the plurality of workloads includes respective workload characteristics. The method also includes determining a workload key for the particular workload based on the respective workload characteristics of the particular workload. The method also includes obtaining a workload history based on determining the workload key and, for each respective cluster of the plurality of clusters, determining a corresponding score associated with executing the particular workload at the respective cluster based on the workload history. The method also includes executing the particular workload at one of the plurality of clusters based on the corresponding score of each respective cluster of the plurality of clusters.