Predictive Bootable Image Writing for Cloud Storage Nodes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In cloud computing environments, the initial instantiation of bootable images is often slow due to the reliance on slower storage elements, which can lead to suboptimal cloud performance and customer experience.

Innovation Solution

Implementing a predictive writing strategy where bootable images are pre-written to fast storage nodes before customer requests, allowing for quick instantiation by distributing the images across multiple storage nodes based on available bandwidth and storage capacity, ensuring that images can be quickly brought online when needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If bootable images are stored on slower storage elements, then storage capacity and cost efficiency are improved, but instantiation speed deteriorates

Engineering Contradiction:

Improveinstantiation speedVSAvoidstorage efficiency

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The system proactively copies bootable images from slower storage elements to faster storage nodes before they are actually requested by customers. This preliminary action ensures that when instantiation is needed, the image is already available on fast storage, eliminating the time penalty of reading from slow storage during the critical instantiation moment.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The storage system is segmented into two distinct layers: slower storage elements for bulk image storage and faster storage nodes for active image delivery. This segmentation allows each layer to perform its optimized function - the slow storage provides capacity and cost efficiency, while the fast storage provides speed for instantiation.

Inventive Principle:

Principle #1Segmentation

2Loss of time

If bootable images are pre-copied to fast storage nodes, then instantiation speed is improved, but storage bandwidth consumption increases

Engineering Contradiction:

Improveinstantiation latencyVSAvoidbandwidth consumption

Core Design Contradiction:

Loss of timeVSLoss of energy

Solution Approach 1:

The system continuously monitors the instantiation frequency of different bootable images and uses this feedback to make intelligent copying decisions. Images that are instantiated frequently are copied to fast storage nodes, while less frequently used images remain on slower storage. This feedback-driven approach optimizes bandwidth usage by only transferring data when it provides measurable performance benefit.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts the copying threshold and frequency based on observed usage patterns. By changing the parameters of when and how often to copy images, the system adapts to varying workloads and optimizes the balance between instantiation speed and bandwidth consumption over time.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS9983888B2Predictive writing of bootable images to storage nodes in a cloud computing environment

Publication Date: 2018.05.29 VERIZON PATENT & LICENSING INC

AI summary

A cloud computing environment may enable customers to choose between a number of possible bootable images. Bootable images, associated with storage nodes in the cloud computing environment, may be predictively copied to other storage nodes. In one implementation, resource usage, with respect to bandwidth and storage capacity, of the storage nodes, may be monitored. A predictive copy may be initiated when the cloud computing environment begins to operate at a level of resource usage that reduces an ability of the cloud computing environment to instantiate new bootable images, as virtual machines, below a threshold.