Task Dispatching via Access Workload Thresholds

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern computer systems face increased workload and response time issues due to high volumes of parallel access requests, leading to resource exhaustion and inefficiencies in scheduling, particularly when the number of tasks accessing a shared resource grows significantly.

Innovation Solution

A method and system for dispatching new tasks to a predefined processing unit when the access workload exceeds a threshold, reducing parallel accesses and alleviating resource exhaustion by implementing a degrading mode that serializes incoming tasks, thereby reducing waiting times and preventing deadlocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the computer system allows parallel access by a large number of users to a resource, then the system can handle more access requests, but the workload increases and response speed deteriorates

Engineering Contradiction:
Improvenumber of access requests handledVSAvoidresponse speed
Core Design Contradiction:
ProductivityVSSpeed

Solution Approach 1:

The system dynamically adjusts the access mode (parallel or serial) based on the current workload conditions. When workload is light, parallel access is enabled to maximize throughput; when workload exceeds thresholds, the system switches to serial access to maintain response speed, making the system adaptive to changing conditions rather than static

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the access parameter (from parallel to serial) based on workload thresholds. By monitoring the number of existing tasks and comparing against predefined thresholds, the system transitions between different access modes to optimize the balance between handling capacity and response speed

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If the computer system continuously receives a huge amount of data access requests, then the workload increases, but the system capacity is exhausted and performance deteriorates

Engineering Contradiction:
Improveamount of data access requestsVSAvoidsystem capacity
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The system takes preliminary action by monitoring workload before it reaches critical levels. By detecting when the number of existing tasks exceeds predefined thresholds, the system proactively switches to serial access mode to prevent resource exhaustion and maintain system capacity, rather than waiting for failure to occur

Inventive Principle:
Principle #9Preliminary anti-action

Solution Approach 2:

The system dynamically transitions between parallel and serial access modes based on real-time workload conditions. This dynamic adjustment allows the system to handle varying quantities of requests while maintaining reliability by switching to a more conservative serial mode when thresholds are exceeded

Inventive Principle:
Principle #15Dynamics

3Productivity

If parallel access is allowed to maximize throughput, then more tasks can access the resource simultaneously, but resource exhaustion and deadlocks occur

Engineering Contradiction:
ImprovethroughputVSAvoidresource exhaustion and deadlocks
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system converts the potentially harmful effect of high workload into a beneficial control mechanism. By using workload thresholds as trigger conditions, the system transforms the risk of resource exhaustion into a proactive switching mechanism that prevents harm while maintaining high throughput when conditions permit

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The workload threshold acts as an intermediary that mediates between parallel and serial access modes. This intermediary condition determines when to switch between access types, preventing direct conflict between the desire for high throughput and the risk of resource exhaustion

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10698785B2Task management based on an access workload
Publication Date: 2020.06.30 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10698785B2 patent drawing
  • US10698785B2 patent drawing
  • US10698785B2 patent drawing

AI summary

A computer-implemented method, a computer program product, and a computer system for parallel task management. A computer system receives a new task that requests to access a resource may be received. In response to an access workload being above a first threshold, the computer system dispatches the new task to at least one predefined processing unit, wherein the access workload may be associated with the resource that is in parallel accessed by a plurality of existing tasks.