Cloud Workload Resource Provisioning Synchronization

Solution Overview

Problem

In cloud computing environments, provisioning multiple computing resources for workloads can lead to inefficiencies due to varying provisioning times, resulting in idle resources and unnecessary costs, as faster resources sit idle while slower ones complete provisioning, wasting physical resources and increasing costs for customers.

Innovation Solution

A method that determines the provisioning time for each computing resource and optimizes the order of provisioning requests to ensure that all resources are ready at substantially the same time, reducing idle time and optimizing resource usage by staggering the start times of slower provisioning operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If provisioning requests are processed in a first-come-first-served manner without considering provisioning times, then the provisioning process is simple to implement, but faster resources sit idle while slower ones complete provisioning, causing resource wastage and increased costs

Engineering Contradiction:

Improveease of implementationVSAvoidresource wastage

Core Design Contradiction:

Ease of manufactureVSLoss of energy

Solution Approach 1:

The system performs preliminary determination of provisioning times for all computing resources before executing provisioning operations. By calculating and comparing provisioning times in advance, the system can stagger the start times of provisioning requests so that all resources complete provisioning simultaneously, eliminating idle time and resource wastage while maintaining implementation simplicity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the start times of provisioning requests based on the determined provisioning times of different computing resources. Instead of a static first-come-first-served approach, the system applies dynamic timing adjustments to ensure synchronized completion, resolving the contradiction between implementation simplicity and resource efficiency

Inventive Principle:

Principle #15Dynamics

2Ease of manufacture

If provisioning requests are processed in a first-come-first-served manner without considering provisioning times, then the provisioning process is simple to implement, but idle time increases causing delayed job execution

Engineering Contradiction:

Improveease of implementationVSAvoididle time

Core Design Contradiction:

Ease of manufactureVSLoss of time

Solution Approach 1:

The system performs preliminary determination of provisioning times for all computing resources before executing provisioning operations. By calculating and comparing provisioning times in advance, the system can stagger the start times of provisioning requests so that all resources complete provisioning simultaneously, eliminating idle time and resource wastage while maintaining implementation simplicity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system ensures continuous useful action by preventing idle periods between provisioning completion and job execution. By synchronizing provisioning completion times, the system eliminates gaps where resources would be idle, maintaining continuous productive operation from provisioning through to job execution

Inventive Principle:

Principle #20Continuity of useful action

3Productivity

If multiple computing resources are provisioned simultaneously without considering their different provisioning times, then resource availability is maximized, but physical resource constraints are exceeded and costs increase

Engineering Contradiction:

Improveresource availabilityVSAvoidphysical resources

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The system performs preliminary determination of provisioning times for all computing resources before executing provisioning operations. By calculating and comparing provisioning times in advance, the system can stagger the start times of provisioning requests so that all resources complete provisioning simultaneously, eliminating idle time and resource wastage while maintaining implementation simplicity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes the timing parameter of provisioning requests based on determined provisioning times. By adjusting start times as a variable parameter, the system optimizes resource utilization to match physical constraints, ensuring that resources are available when needed without exceeding capacity limits or incurring unnecessary costs

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10372501B2Provisioning of computing resources for a workload

Publication Date: 2019.08.06 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Provisioning of computing resources for a workload in a networked computing environment. A method receives a set of provisioning requests in a computer data structure of a networked computing environment, wherein the provisioning requests relate to a workload request. The method further identifies a set of provisioning operations for computing resources in the networked computing environment to perform the provisioning requests. The method determines, for each provisioning operation of the provisioning operations, a respective provisioning time, the respective provisioning time being an amount of time for a particular computing resource to become prepared and equipped to perform a job. The determining the respective provisioning time for each provisioning operation provides a plurality of determined provisioning times. The method also determines an order for the provisioning requests based on the plurality of determined provisioning times, such that the provisioning operations complete at substantially a same time.