Operation Pool Scheduling for Cloud Object Storage Throttling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based storage systems face inefficiencies due to throttling errors when handling high volumes of requests, leading to resource misallocation between clients and the cloud-based system, as they rely on rudimentary retry logic to manage access requests.
Innovation Solution
Implementing a server that acts as an intermediary between clients and object storage, utilizing an operation pool to manage operation-specific queues and schedule data operations efficiently, allowing for concurrent execution of sub-operations based on bandwidth limits, thereby improving data operation performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cloud-based storage systems handle high volumes of requests simultaneously, then service coverage and client satisfaction improve, but throttling errors increase and system reliability deteriorates
Solution Approach 1:
The patent implements preliminary action by pre-establishing operation pools and scheduling queues before requests arrive. The system proactively manages request queuing and scheduling, rather than reacting to throttling errors after they occur. This allows the system to handle high request volumes by distributing them across multiple operation pools and queues, preventing overload and maintaining reliability.
Solution Approach 2:
The patent introduces an intermediary scheduling layer between clients and the cloud-based storage system. This scheduling component acts as a mediator that buffers and regulates request flow, preventing direct transmission of overwhelming request volumes to the storage system. The intermediary manages operation pools and queues, smoothing out traffic spikes and preventing throttling errors while maintaining high service coverage.
2Reliability
If clients use retry logic to handle throttling errors, then request completion rate improves, but resource allocation efficiency deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where the scheduling system continuously monitors operation pool status, queue lengths, and system capacity. Based on this feedback, the system dynamically adjusts request scheduling, prioritizing operations and managing queues to ensure completion without requiring aggressive retry logic. This feedback-driven approach completes requests efficiently while avoiding wasteful resource consumption from blind retry attempts.
Solution Approach 2:
The patent applies dynamics by making the scheduling system adaptive and flexible rather than static. The operation pools and queues dynamically adjust their capacity and prioritization based on current system conditions and request patterns. This dynamic scheduling ensures high request completion rates while optimizing resource allocation, eliminating the need for inefficient fixed retry logic.
3Adaptability or versatility
If the cloud-based storage system increases request handling capacity, then service scalability improves, but resource management complexity increases
Solution Approach 1:
The patent segments the request handling system into multiple operation pools and scheduling queues, each managing specific types or priorities of operations. This segmentation allows the system to scale independently for different operation types without managing a single monolithic complex system. Each segment can be optimized and managed separately, reducing overall resource management complexity while maintaining high scalability.
Solution Approach 2:
The patent creates a universal scheduling framework that handles multiple types of operations through a common architecture. The operation pools and queues can accommodate various request types uniformly, allowing the system to scale services without creating separate management mechanisms for each operation type. This multi-functional approach simplifies resource management while maintaining adaptability and scalability.
Data Source
AI summary
Described is a system for cloud-based scheduling of data operations to an object storage. The system may include a server that acts as an intermediary between a client device and an object storage that stores client data. Requests by the client device to manage data on the object storage may be routed through the server. In response, the server may execute corresponding data operations directly on the object storage. The server may manage bandwidth limits for operation requests to an object storage by maintaining a specialized operation pool. The operation pool may be configured to manage operation-specific queues, and these operation-specific queues may include sub-operations (or threads) that are executed on the object storage. Sub-operations of various types of operations may be initiated for concurrent execution on the object storage based on scheduling information derived from the operation pool.