On-Demand Compute Integration for Peak Load Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current resource management systems in high-performance computing environments face challenges in efficiently utilizing on-demand compute resources to handle peak loads without unnecessary infrastructure costs, as they lack effective solutions for communication and connectivity with remote high-performance computing centers.
Innovation Solution
A method and system that integrate on-demand compute environments with local environments, allowing for the automatic provisioning and allocation of resources based on workload requirements, using management modules to manage and provision resources, ensuring compliance with data management policies, security, and quality of service, while reducing infrastructure costs by only utilizing extra processing power during peak demand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If organizations build infrastructure to accommodate peak demand, then service availability during peak loads is improved, but infrastructure costs increase
Solution Approach 1:
The system dynamically provisions compute resources by transitioning between local and remote environment based on real-time workload conditions. The resource manager automatically scales resource allocation up or down, allowing the infrastructure to adapt flexibly to varying demand without over-provisioning for peak loads, thus resolving the contradiction between reliability and cost.
Solution Approach 2:
The architecture enables a single infrastructure to serve multiple functions: local compute environment handles normal operations while remote on-demand environment provides peak load capacity. This multi-functionality allows the system to maintain service availability without permanently maintaining expensive peak-capacity infrastructure, as the same physical resources serve different purposes at different times.
2Quantity of substance
If organizations use local compute environment only, then infrastructure costs are reduced, but ability to handle peak loads is worsened
Solution Approach 1:
The resource manager acts as an intermediary between the local compute environment and remote on-demand environment. It transparently manages resource allocation, automatically routing workload to remote resources when peak loads are detected, thus enabling peak load handling capability without requiring permanent expensive infrastructure.
Solution Approach 2:
The compute environment is segmented into local and remote components, each optimized for different operational scenarios. The local environment handles routine workloads cost-effectively, while the remote environment provides burst capacity for peak loads. This segmentation allows the system to achieve peak load handling capability without uniformly expanding the entire infrastructure.
3Productivity
If on-demand compute environment is integrated, then resource utilization efficiency is improved, but system complexity increases
Solution Approach 1:
The resource manager implements self-service automation, automatically monitoring workload conditions, determining when remote resources are needed, and provisioning them without manual intervention. This self-service capability improves resource utilization efficiency by dynamically allocating resources based on actual demand, while the automation masks much of the underlying system complexity from users.
Solution Approach 2:
The system incorporates continuous feedback loops where the resource manager monitors local environment workload, compares it against thresholds, and automatically triggers resource provisioning or deprovisioning actions. This feedback mechanism enables efficient resource utilization through dynamic adaptation, while the automated control loops manage system complexity by providing clear cause-effect relationships.
Data Source
AI summary
Disclosed are a system and method of integrating an on-demand compute environment into a local compute environment. The method includes receiving a request from an administrator to integrate an on-demand compute environment into a local compute environment and, in response to the request, automatically integrating local compute environment information with on-demand compute environment information to make available resources from the on-demand compute environment to requesters of resources in the local compute environment such that policies of the local environment are maintained for workload that consumes on-demand compute resources.