Reactive Throttling of Heterogeneous Migration Sessions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for throttling migration sessions in virtualized cloud computing data centers are limited to homogeneous sessions and cannot effectively manage congestion caused by heterogeneous migration sessions, leading to excessive I/O latency and potential service level agreement violations.
Innovation Solution
The use of virtual-to-physical resource mapping information to identify and throttle heterogeneous migration sessions contributing to bottlenecks, allowing for targeted reduction of traffic congestion by adjusting the migration rate of each session based on its function and contribution to the bottleneck.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple heterogeneous migration sessions are initiated concurrently to improve resource utilization, then productivity is improved, but I/O latency increases and service level agreements may be violated
Solution Approach 1:
The patent dynamically adjusts the migration rate parameter of each heterogeneous migration session based on real-time bottleneck detection and contribution analysis. When a resource bottleneck is detected, the system modifies the migration rate parameter individually for each session that contributes to the bottleneck, thereby reducing I/O latency while maintaining overall productivity through selective rather than uniform throttling.
2Loss of time
If traditional throttling methods are applied to reduce migration transfer rates, then I/O latency is reduced, but the methods are limited to homogeneous sessions and cannot effectively manage heterogeneous migration sessions
Solution Approach 1:
The patent segments the throttling control by identifying individual migration sessions and their specific contributions to resource bottlenecks. Rather than applying uniform throttling to all sessions, the system divides the control into session-specific segments, analyzing each session's I/O patterns and contribution to bottlenecks separately, thereby enabling effective management of heterogeneous migration sessions with different characteristics.
Solution Approach 2:
The patent applies different throttling strategies and parameters to different migration sessions based on their specific characteristics and contribution to bottlenecks. Each session receives localized quality control tailored to its function type (VM, file, volume migration) and its specific impact on resource utilization, rather than applying a one-size-fits-all approach.
3Productivity
If migration sessions are throttled to alleviate network congestion, then resource utilization is improved, but the complexity of managing and monitoring multiple heterogeneous sessions increases
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors resource utilization and bottleneck conditions, then automatically adjusts migration rates accordingly. The system detects bottlenecks in real-time, analyzes which sessions contribute to them, and applies appropriate throttling without requiring complex manual intervention. This automated feedback loop simplifies management despite the heterogeneity of sessions.
Solution Approach 2:
The patent enables migration sessions to effectively self-regulate their rates based on system conditions. The system provides the framework for automatic detection and adjustment, where each session's contribution to bottlenecks is automatically analyzed and appropriate throttling is applied without requiring external complex management intervention for each individual session.
Data Source
AI summary
An approach for reactive throttling of heterogeneous migration sessions running concurrently in a virtualized cloud environment is disclosed. The heterogeneous migration sessions are characterized by several attributes such as the type of function performed by the migration session and the value that is attained by performing the migration session. An input/output (I/O) path to any resource in the virtualized cloud environment computing that is experiencing excessive resource congestion is identified. Bottlenecks along the I/O path are then determined. Virtual-to-physical resource mapping information of resources in the virtualized cloud computing environment are used to obtain a list of the heterogeneous migration sessions contributing to the bottlenecks along the identified I/O path.