Grid Computing Resource Allocation for Concurrent Batch and Interactive Sessions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional grid computing environments face inefficiencies in resource allocation, particularly during off-peak times when thin client servers may be unutilized, as they are not optimized to concurrently execute both batch jobs and interactive sessions, leading to underutilization of resources.
Innovation Solution
A method is introduced to determine and allocate computer resources for simultaneously executing batch jobs and interactive sessions within a grid computing environment, utilizing a resource management server that schedules and allocates resources across multiple compute nodes, ensuring efficient use of resources by optimizing CPU, memory, and network bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If grid computing environments are optimized for batch jobs, then computational intensive applications can be efficiently executed, but interactive sessions cannot be supported and servers remain unutilized during off-peak times
Solution Approach 1:
The grid computing environment is designed to perform multiple functions by allowing the same compute nodes and resources to handle both batch jobs and interactive sessions. The system dynamically allocates resources based on workload type, enabling servers to serve computational intensive batch processing during peak times and interactive thin client sessions during off-peak times, thus achieving multi-functionality without requiring separate infrastructure
2Ease of operation
If servers are dedicated to interactive sessions, then user applications can run smoothly, but resources remain unutilized during off-peak times
Solution Approach 1:
The system implements dynamic resource allocation where compute nodes can transition between handling interactive sessions and batch jobs based on real-time workload demands. During off-peak hours, servers dynamically accept and execute batch jobs alongside or instead of interactive sessions, optimizing resource utilization without compromising user application performance when needed
3Productivity
If resources are allocated for batch jobs, then computational power is efficiently utilized, but interactive session support is reduced
Solution Approach 1:
The system allows compute nodes to partially dedicate capacity to both batch jobs and interactive sessions simultaneously. Rather than fully allocating resources to one function or the other, the system enables partial participation in both workloads, allowing batch jobs to utilize available computational power while interactive sessions receive sufficient resources to maintain operational ease
Data Source
AI summary
A method of allocating resources in a grid computing environment is provided. The method includes allocating resources for concurrently executing a batch job and an interactive session.