Predictive Migration Scheduler for Virtualized Hosts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data migration techniques in virtualized environments often result in inefficient resource utilization due to idle time caused by preset migration schedules, as they do not accurately predict the completion time of process migrations and the end times of processes on target hosts, leading to suboptimal use of resources and increased energy consumption.
Innovation Solution
The implementation of predictive analysis techniques that analyze historical data to determine the optimal time for migrating processes, ensuring that migrations are completed as soon as or after the processes on target hosts have finished, thereby minimizing idle time and improving resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If preset migration schedules are used for process migration, then migration can be performed in a controlled manner, but idle time increases and resource utilization decreases
Solution Approach 1:
The patent applies dynamics by transitioning from static preset migration schedules to dynamic predictive scheduling. The system uses historical data and predictive analysis to adapt migration timing based on actual process behavior patterns, allowing the migration schedule to evolve and optimize itself continuously rather than following fixed predetermined times
Solution Approach 2:
The patent implements feedback mechanisms by analyzing historical migration data and process execution data to continuously improve prediction accuracy. The system monitors actual migration completion times and process end times, uses this feedback to refine predictive models, and adjusts future migration schedules accordingly to minimize idle time while maintaining controlled migration execution
2Measurement precision
If predictive analysis techniques are implemented to predict migration completion time and process end time, then migration scheduling precision is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by collecting and storing historical migration data and process execution data in advance. This pre-collected data is used to train predictive models before actual migration scheduling decisions are made, allowing the system to make accurate predictions without adding real-time complexity to the migration execution process
Solution Approach 2:
The patent uses copying by creating predictive models that replicate the behavior patterns observed in historical data. Instead of directly analyzing complex real-time system states, the system copies past behavior patterns into predictive models that can be applied repeatedly to forecast migration completion times and process end times with high accuracy
3Ease of operation
If migration is scheduled without accurate prediction of completion times, then scheduling is simple, but idle time increases and energy consumption increases
Solution Approach 1:
The patent applies self-service by enabling the migration scheduling system to automatically predict completion times and optimize schedules without requiring manual intervention. The predictive analysis system autonomously analyzes historical data, generates predictions, and determines optimal migration timings, eliminating the need for complex manual scheduling while reducing idle time
Data Source
AI summary
Systems, methods, and machine-readable instructions stored on machine-readable media are disclosed for selecting, based on an analysis of a first process executing on a first host, at least one of a plurality of other hosts to which to migrate the first process, the selecting being further based on an analysis of the plurality of the other hosts and an analysis of processes executing on the plurality of the other hosts. At least one predictive analysis technique is used to predict an amount of time to complete migrating the first process to the selected at least one of the plurality of other hosts and an end time of the second process. In response to determining that a current time incremented by the predicted amount of time to complete migrating the first process is later than or equal to the predicted end time of the second process, a migration time at which to migrate the first process from the first host to the selected at least one of the plurality of other hosts is scheduled. At the scheduled migration time, the migration of the first process from the first host to the selected at least one of the plurality of other hosts is performed.