Dynamic Resource Allocation in Shared Computing Pools
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing a shared pool of configurable computing resources efficiently becomes challenging as data volume increases, requiring dynamic resource allocation and balancing to maintain performance and availability.
Innovation Solution
Implementing capacity-on-demand technology and live migration to dynamically assign and reassign resources based on profile data and policies, allowing for efficient distribution and activation of resources to achieve threshold utilization, such as through mobile resource distribution or virtual machine migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resources are statically allocated to hosts, then resource utilization is predictable and stable, but resource efficiency decreases when demand fluctuates
Solution Approach 1:
The patent implements dynamic resource allocation where computing resources are not permanently bound to specific hosts but are allocated based on real-time demand. The system monitors resource utilization metrics and dynamically assigns or reassigns resources between hosts, allowing the allocation to adapt to changing workload conditions while maintaining stability through controlled transition mechanisms
2Reliability
If computing resources are increased to handle peak demand, then service availability is maintained, but resource waste occurs during low-utilization periods
Solution Approach 1:
The patent creates a shared pool of computing resources that can serve multiple hosts and workloads universally. Instead of dedicating resources to single hosts, the same resource pool is shared across multiple computational tasks, allowing resources to be dynamically allocated to different hosts based on current demand, thus eliminating waste during low-utilization periods while maintaining availability during peaks
3Productivity
If resources are dynamically reassigned between hosts, then resource efficiency improves, but system complexity increases
Solution Approach 1:
The patent introduces a resource pool manager as an intermediary component that handles the complexity of dynamic resource allocation. This manager sits between the physical resources and the hosts, abstracting the complex reassignment logic into managed operations. The manager handles resource tracking, allocation decisions, and reassignment coordination, thereby reducing the complexity burden on individual hosts while enabling efficient dynamic resource management
4Reliability
If live migration is used to balance load, then resource balancing is achieved, but migration time and performance overhead occur
Solution Approach 1:
The patent implements preliminary resource allocation and pre-positioning strategies where resources are proactively assigned to appropriate hosts based on predicted or current workload patterns before actual demand peaks occur. By anticipating load imbalances and pre-allocating resources accordingly, the system reduces the need for urgent live migration operations, thereby minimizing migration time and performance overhead while maintaining effective load balancing
Data Source
AI summary
Disclosed aspects manage a shared pool of configurable computing resources. A triggering event is detected. The triggering event may be related to a resource utilization on a host of the shared pool of configurable computing resources. Based on a set of profile data, it is determined to perform an event response. The event response includes initiating a resource action or initiating an asset action. The resource action may include distributing a set of dynamically-assigned resources. The asset action corresponds to a set of assets (e.g., migrating a set of virtual machines). To change the resource utilization on the host, the event response is performed.