Shared Resource Management via Obstructing Stream Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based systems face latency and reduced throughput due to sudden changes in data streams, such as increased byte rates or processing times, which can violate service-level agreements and impact performance for multiple tenants sharing resources.
Innovation Solution
Identify obstructing streams by computing statistical values like byte-rate and processing-time values, allocate new shared computer resources, and create new computing processes to isolate and process these streams separately, reducing the load on other streams and optimizing resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If shared computer resources are dynamically provisioned and de-provisioned to match actual resource needs, then resource allocation efficiency is improved, but system latency increases and throughput decreases due to sudden changes in data streams
Solution Approach 1:
The system performs preliminary actions by computing statistical values (byte-rate, processing-time) and identifying obstructing streams before they cause significant performance degradation. By detecting streams that exceed thresholds in advance, the system can proactively allocate resources or redirect traffic, preventing latency and throughput issues before they occur.
Solution Approach 2:
The system segments data streams by identifying and separating obstructing streams from normal streams based on statistical analysis. Obstructing streams are routed through different computing processes or allocated dedicated resources, while normal streams continue using shared resources. This segmentation prevents high-volume or slow-processing streams from degrading overall system performance.
2Reliability
If more shared computer resources are allocated to handle obstructing streams, then service quality for all tenants is maintained, but resource cost increases significantly
Solution Approach 1:
The system applies local quality by providing differentiated resource allocation based on stream characteristics. Obstructing streams receive enhanced resources or dedicated processing paths, while normal streams continue using standard shared resources. This targeted approach maintains service quality for problematic streams without unnecessarily increasing resources for all streams, thereby controlling costs.
Solution Approach 2:
The system introduces an intermediary layer (stream management component) that analyzes statistical values and mediates resource allocation. This intermediary identifies obstructing streams and routes them through appropriate computing processes, acting as a buffer between raw data streams and resource allocation decisions. This mediation enables efficient resource usage by directing resources only where needed.
3Measurement precision
If statistical analysis is performed on all streams to identify obstructing streams, then stream performance monitoring is improved, but processing overhead increases
Solution Approach 1:
The system applies partial action by performing statistical analysis selectively rather than uniformly on all streams. It computes statistical values (byte-rate, processing-time) and compares them against thresholds to identify only those streams that exhibit obstructing behavior. This selective analysis reduces processing overhead compared to continuous monitoring of all streams, while maintaining sufficient measurement precision to detect performance issues.
Data Source
AI summary
A system for managing a plurality of shared computer resources, comprising at least one hardware processor adapted for: in at least one of a plurality of management iterations: computing statistical values each associated with one of a plurality of streams, where the streams are input to a plurality of computing processes which are used for processing the plurality of streams, executed using at least one of a plurality of shared computer resources; identifying according to an analysis of the statistical values at least one obstructing stream; allocating at least one new shared computer resource; creating a plurality of new computing processes which execute using the at least one new shared computer resource; and instructing inputting the at least one obstructing stream to the plurality of new computing processes instead of to the plurality of computing processes while the plurality of streams are inputted to the plurality of computing processes.