Transaction Scheduler for Cooperative Partition Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data processing systems lack the ability to run multiple partitions simultaneously within a single time window without disrupting hard time/space partition scheduling.
Innovation Solution
A transaction scheduler is introduced, which determines whether a selected partition is part of a transaction, identifies the tail partition, and schedules it to run in the time window, enabling cooperative scheduling of multiple partitions in a single time frame through a method involving transaction start and terminate system calls.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional partition scheduling is used to allocate resources to partitions, then resource allocation is simple and clear, but multiple partitions cannot run simultaneously in a single time window
Solution Approach 1:
The invention segments the time window into multiple sub-time windows, allowing different partitions to execute in different sub-windows. This segmentation enables concurrent execution of multiple partitions while maintaining clear resource allocation boundaries, resolving the contradiction between productivity and device complexity.
Solution Approach 2:
The scheduling system dynamically adjusts partition execution by determining whether a partition should run in the current time window based on transaction state. The system can dynamically extend transactions across multiple time windows and coordinate multiple partitions to run cooperatively, enabling concurrent execution without breaking time/space partition scheduling guarantees.
2Productivity
If multiple partitions are scheduled to run in a single time window, then resource utilization improves, but hard time/space partition scheduling guarantees are broken
Solution Approach 1:
By segmenting the time window into sub-time windows and assigning different partitions to different sub-windows, the system achieves improved resource utilization while maintaining the isolation and scheduling guarantees that each partition expects. The segmentation ensures that partitions do not interfere with each other's time/space requirements.
Solution Approach 2:
The invention introduces a transaction coordinator as an intermediary that manages multiple partitions participating in a transaction. The coordinator ensures that all partitions complete their operations within the extended time window while maintaining scheduling guarantees, acting as a mediator between the need for concurrent execution and the requirement for reliable time/space partitioning.
3Adaptability or versatility
If transactions are extended dynamically within a time window, then flexibility and adaptability improve, but scheduling complexity increases
Solution Approach 1:
The system dynamically extends transactions by allowing partitions to participate in transactions that span multiple time windows. The transaction state is tracked and updated dynamically, enabling flexible adaptation to different transaction requirements while the underlying scheduling mechanism maintains structure through time window segmentation.
Solution Approach 2:
The scheduling system uses feedback from transaction state to determine whether partitions should run in the current time window. The system monitors transaction progress and dynamically adjusts scheduling decisions based on this feedback, enabling flexible transaction extension while managing complexity through structured feedback mechanisms.
Data Source
AI summary
Described is a method for determining whether a first partition selected to run in a time window is a member of a transaction, the transaction including at least one further partition, determining a tail partition from among the at least one further partition and scheduling the tail partition to run in the time window. A transaction scheduler having a transaction determination module determining whether a first partition selected to run in a time window is a member of a transaction, the transaction including at least one further partition, a tail determination module determining a tail partition from among the at least one further partition and a scheduling module scheduling the tail partition to run in the time window.