Depleting Chips for IT Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches fail to adequately ration computational resources among competing parties, leading to suboptimal utilization and increased costs in IT service delivery centers, as they do not effectively balance resource allocation to meet the needs of individual parties without hindering others.
Innovation Solution
The implementation of discrete, depleting chips, which are allocated to parties for a fixed period, allowing them to bid on computational resources and debit the chip amount upon usage, with the option to replenish if depleted, facilitating balanced resource allocation and utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing resource allocation approaches are used, then resource distribution occurs among competing parties, but resource utilization remains suboptimal and costs increase due to inadequate rationing
Solution Approach 1:
The patent introduces a broker as an intermediary entity that mediates resource allocation between multiple competing parties. The broker receives resource requests, evaluates them against service level characteristics, and allocates resources optimally. This intermediary mechanism enables better resource utilization by coordinating competing demands rather than allowing direct party-to-party competition, thereby reducing waste and costs while improving overall productivity.
2Reliability
If resources are allocated to meet individual party needs, then service level characteristics are satisfied, but other parties' needs may be hindered due to lack of rationing
Solution Approach 1:
The patent implements dynamic resource allocation where the broker continuously evaluates resource requests against evolving service level characteristics and available capacity. The system adapts allocation decisions in real-time based on current system state, party priorities, and resource availability. This dynamic approach ensures service level requirements are met while maintaining flexibility to respond to changing conditions and balance competing party needs effectively.
Solution Approach 2:
The system changes allocation parameters dynamically based on service level characteristics and resource availability. The broker adjusts allocation thresholds, priority weights, and resource quotas as parameters to balance individual party service level requirements with overall system resource utilization. This parameter adjustment mechanism enables the system to satisfy reliability requirements while adapting to different operational conditions and party needs.
3Quantity of substance
If multiple parties compete for common resources, then resource distribution occurs, but adequate rationing is not achieved leading to suboptimal utilization
Solution Approach 1:
The patent implements a feedback mechanism where the broker continuously monitors resource allocation outcomes, service level characteristic fulfillment, and party requests. This feedback loop enables the broker to learn from allocation results and adjust future allocation decisions to improve resource utilization. The system uses feedback from resource consumption patterns and service level measurements to optimize distribution quantities and prevent both over-allocation and under-utilization, thereby improving overall productivity while maintaining adequate resource distribution.
Data Source
AI summary
The present invention provides discrete, depleting chips for allocating computational resources for obtaining desired service level characteristics, wherein discrete chips deplete from a maximum allocated amount but may, in an optional implementation, be allowed to be replenished through the purchase of additional chips. A number of chips are assigned to a requestor/party, known as a business unit (BU), which could be a department, or group providing like-functionality services. In one implementation, the chips themselves could represent base monetary units integrated over time.