Microservice Storage Allocation via Predictive Categorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage allocation methods for microservices are inefficient due to unpredictable storage behavior and lack of optimization in communication between microservice orchestrators and block storage controllers, leading to suboptimal storage allocation and garbage collection.
Innovation Solution
A computer-implemented method and system that categorizes microservice containers based on predicted storage behavior using categorization analytics, which analyzes input/output operation patterns to optimize storage allocation and garbage collection by providing categorization details to the storage system, employing machine learning techniques to cluster patterns and adapt categorizations dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If storage allocation is performed without categorization analytics, then the storage system is simpler to operate, but storage allocation efficiency and garbage collection performance deteriorate
Solution Approach 1:
A categorization analytics component is introduced as an intermediary between the microservice orchestrator and the storage system. This component analyzes input/output operation patterns, determines storage behavior categories, and provides categorization information to the storage system, enabling optimized storage allocation and garbage collection without requiring complex changes to the core storage architecture
Solution Approach 2:
The system performs preliminary analysis of input/output operation patterns before actual storage allocation occurs. By analyzing patterns in advance and determining storage behavior categories beforehand, the system can prepare appropriate storage allocation strategies ahead of time, improving efficiency without adding complexity during the allocation process itself
2Loss of time
If garbage collection is performed without category-based optimization, then the garbage collection process is simpler, but garbage collection overheads and time increase
Solution Approach 1:
The garbage collection process is optimized by applying different strategies to different data categories. Instead of uniform treatment, the system identifies specific categories of input/output operations and applies tailored garbage collection approaches for each, reducing overall garbage collection time and overhead while maintaining manageable complexity through localized optimization
3Reliability
If storage allocation does not consider input/output operation patterns, then the storage system requires less analysis capability, but storage allocation becomes suboptimal
Solution Approach 1:
The system continuously monitors and analyzes input/output operation patterns from microservice containers and uses this feedback to determine storage behavior categories. This feedback loop enables the system to adapt storage allocation decisions based on actual observed behavior, improving allocation optimality while managing detection complexity through systematic pattern analysis
Data Source
AI summary
Method and system are provided for storage allocation enhancement of microservices. A method carried out at a microservice orchestrator, includes: categorizing a microservice container, wherein the categorization defines a predicted storage behavior of the microservice container input/output operations; and providing the categorization in association with the microservice container input/output operations to a storage system for use in storage allocation of the input/output operations. A method at a storage controller includes: receiving microservice container input/output operations with an associated categorization, wherein the categorization defines a predicted storage behavior of the microservice container input/output operations; and using the associated categorization for optimizing storage allocation for the input/output operations and/or optimizing garbage collection performance.