Virtual Private Cluster for On-Demand Compute Workload Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current on-demand computing environments face challenges in efficiently and flexibly utilizing resources due to inflexibility in hardware and resource manager environments, leading to delays and increased costs for users, as they require alignment with specific hardware and operating systems, making it burdensome for customers to access and utilize peak compute resources.
Innovation Solution
The system enables the creation of a virtual private cluster within an on-demand center by reserving and provisioning resources to duplicate a local compute environment, allowing for advanced resource reservations and dynamic workload transfer, thereby optimizing resource utilization and reducing costs through intelligent workload management and data staging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If users align their local compute environment with specific hardware and operating systems required by on-demand centers, then access to on-demand resources is enabled, but infrastructure change costs and complexity increase significantly
Solution Approach 1:
The patent creates a virtual copy of the user's local compute environment (including operating system, hardware configuration, and software stack) within the on-demand center. This copy allows workload transfer without requiring the user to physically modify their local infrastructure, resolving the contradiction by enabling access while avoiding infrastructure changes
Solution Approach 2:
The patent introduces a virtualization layer and workload management system as an intermediary between the user's local environment and the on-demand center's hardware. This intermediary translates and adapts the workload, allowing compatibility without direct alignment between user hardware and on-demand hardware specifications
2Productivity
If on-demand centers require specific hardware and operating system alignment, then resource management is simplified, but access delays increase due to infrastructure changes
Solution Approach 1:
The patent performs preliminary virtualization and environment setup in advance, creating ready-to-use virtual copies of compute environments before workload transfer is needed. This preliminary action eliminates access delays by having pre-configured resources available immediately when workload transfer is initiated
Solution Approach 2:
By creating virtual copies of the local environment in advance and maintaining them in a ready state at the on-demand center, the system enables immediate workload transfer without requiring time-consuming infrastructure changes or alignment processes
3Productivity
If users transfer workload to on-demand centers, then peak compute resource utilization is improved, but compatibility issues arise due to hardware and software environment mismatches
Solution Approach 1:
The patent creates accurate virtual copies of the user's local compute environment including operating system, hardware abstraction layer, and software stack. This copying ensures that the workload runs in an identical environment at the on-demand center, guaranteeing compatibility and reliability while enabling peak resource utilization
Solution Approach 2:
The virtualization platform provides a universal execution environment that can accommodate multiple different hardware configurations and operating systems. This universality allows workloads from diverse environments to be transferred and executed reliably on the on-demand center's hardware without compatibility issues
4Stability of the object's composition
If on-demand centers implement strict hardware and software alignment requirements, then system stability is improved, but ease of operation deteriorates due to burdensome configuration changes
Solution Approach 1:
The system automatically creates virtual copies of the user's existing environment configuration, preserving the stable and proven setup locally while replicating it virtually at the on-demand center. This copying approach maintains system stability by using the same configuration while eliminating the need for users to perform burdensome configuration changes
Solution Approach 2:
The workload management system automatically handles the virtualization, environment replication, and configuration matching processes without requiring user intervention. This self-service capability maintains system stability through automated configuration management while dramatically improving ease of operation by removing burdensome manual setup requirements
Data Source
AI summary
Disclosed are systems, methods and computer-readable media for controlling and managing the identification and provisioning of resources within an on-demand center as well as the transfer of workload to the provisioned resources. One aspect involves creating a virtual private cluster within the on-demand center for the particular workload from a local environment. A method of managing resources between a local compute environment and an on-demand environment includes detecting an event associated with a local compute environment and based on the detected event, identifying information about the local environment, establishing communication with an on-demand compute environment and transmitting the information about the local environment to the on-demand compute environment, provisioning resources within the on-demand compute environment to substantially duplicate the local environment and transferring workload from the local-environment to the on-demand compute environment. The event can be a threshold or a triggering event within or outside of the local environment.