Dynamic Structural Variation Detection Task Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current structural variation detection methods in human whole-genome re-sequencing are inefficient due to uneven task allocation, leading to idle computer resources and prolonged testing durations, as they do not effectively utilize concurrent processing across different loci and chromosome segments.

Innovation Solution

A method that divides test sequences into portions based on chromosome loci, allocates tasks to detection apparatuses, monitors task completion, and dynamically adjusts by killing incomplete tasks when a preset proportion threshold is reached, allowing for redivision and continued processing to maximize resource utilization and reduce detection time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a simple concurrent calculation scheme divides chromosomes into N portions for parallel processing, then computer resources are utilized more effectively, but processing duration varies significantly across different loci causing idle resources and extended total testing time

Engineering Contradiction:
Improvecomputer resource utilizationVSAvoidtotal testing duration
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent implements dynamic task allocation by continuously monitoring the completion status of detection tasks across multiple apparatuses. When a preset proportion of tasks are completed, the system dynamically adjusts by sending kill instructions to incomplete tasks and reallocating resources, transforming the static division into an adaptive process that responds to actual processing speeds and resource availability

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system establishes a feedback mechanism where the management apparatus continuously detects the completion situations of detection tasks from multiple detection apparatuses. This feedback loop enables the system to monitor processing progress in real-time and make informed decisions about task allocation and resource reallocation, ensuring optimal utilization of computational resources throughout the detection process

Inventive Principle:
Principle #23Feedback

2Speed

If chromosome segments are processed concurrently with fixed task allocation, then processing speed increases, but uneven processing durations across segments cause resource idle time

Engineering Contradiction:
Improveprocessing speedVSAvoididle computer resources
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

The patent transforms fixed task allocation into a dynamic system that adjusts task distribution based on real-time completion status. By monitoring which detection apparatuses have completed their tasks and which are still processing, the system dynamically reallocates resources to maintain balanced utilization across all apparatuses, preventing idle time while preserving concurrent processing speed

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameter of task allocation from a static fixed distribution to a dynamic distribution that varies based on completion proportions. When the proportion of completed tasks reaches a preset threshold, the system modifies task parameters by killing incomplete tasks and reallocating resources, thereby adjusting resource utilization parameters to eliminate idle time while maintaining high processing speed

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10950328B2Method, apparatus and system for detecting structural variations
Publication Date: 2021.03.16 BEIJING BAIDU NETCOM SCI & TECH CO LTD
  • US10950328B2 patent drawing
  • US10950328B2 patent drawing
  • US10950328B2 patent drawing

AI summary

A method, apparatus and system for detecting structural variations is provided. A management apparatus divides a test sequence according to loci of chromosomes to obtain at least two portions of detection tasks, sends respective detection tasks to respective detection apparatuses and activates the respective detection tasks; detects detection task completion situations of detection apparatuses and determines whether the number of uncompleted tasks is reduced to a preset proportion threshold of a total number of tasks; when the number is reduced to a preset proportion threshold of the total number of tasks, the management apparatus sends, to detection apparatuses that have not yet completed detection tasks, an instruction message to kill uncompleted detection tasks; the management apparatus further divides the uncompleted detection tasks into at least two portions, sends respective detection tasks to respective detection apparatuses, and activates said respective detection apparatuses to continue to perform detection of structural variations.