Computing Resource Allocation via Statistical Queue Length Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computing network environments, dynamic job arrival patterns lead to ineffective planning and allocation of computing resources, resulting in overloading or underutilization, which can affect compliance with service level agreements (SLAs) due to the lack of efficient resource management.
Innovation Solution
A system and method that determine the number of computing resources required by receiving jobs and associated variables such as mean processing time, standard deviation, service level agreement, and confidence level, computing a queue length, and using this information to calculate the necessary resources through a ceiling function, ensuring effective resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If computing resources are allocated without effective planning due to dynamic job arrival, then the system can accept dynamic job loads, but computing resources become overloaded or idle leading to ineffective utilization
Solution Approach 1:
The system performs preliminary calculation of the number of computing resources needed before job execution begins. By using the received variables (mean processing time, standard deviation, service level agreement, confidence level) to compute the required resources in advance, the system prepares an optimal allocation plan that prevents both overload and idle states during dynamic job arrival scenarios
Solution Approach 2:
The system uses statistical feedback from historical job data (mean processing time μ and standard deviation σ) to continuously adjust resource allocation decisions. This feedback mechanism allows the system to adapt to dynamic job patterns while maintaining optimal resource utilization by comparing actual performance against expected parameters
2Reliability
If more computing resources are allocated to ensure SLA compliance, then job execution within time limits is improved, but resource cost and system complexity increase
Solution Approach 1:
The system changes the parameters of resource allocation from static to dynamic by incorporating statistical parameters (mean μ, standard deviation σ) and service level parameters (confidence level 1-α, service level agreement α). These parameter changes enable precise calculation of optimal resource numbers that ensure SLA compliance without excessive allocation
Solution Approach 2:
The patent introduces mathematical formulas and statistical models as intermediaries between job arrival patterns and resource allocation decisions. The queue length calculation formula acts as a mediator that translates complex dynamic job characteristics into concrete resource allocation numbers, simplifying the decision-making process while ensuring SLA compliance
3Productivity
If the number of computing resources is precisely determined using statistical parameters, then resource allocation efficiency is improved, but calculation complexity and data requirements increase
Solution Approach 1:
The patent creates a universal resource allocation formula that serves multiple functions: it calculates optimal resource numbers, ensures SLA compliance, and adapts to different job types simultaneously. The single formula √k=−σγ±√(σ2γ2+4T μ)/2μ integrates multiple considerations (processing time, variability, service level requirements) into one comprehensive calculation, improving efficiency without proportionally increasing complexity
Data Source
AI summary
Method and system for determining the number of computing resources for executing a plurality of jobs. In accordance with the method and the system, the plurality of jobs along with a plurality of variables associated to the plurality of jobs may be received. The plurality of variables may include a mean processing time (μ), a standard deviation (σ), a service level agreement (α), and a confidence level (1-α). After the receipt of the plurality of variables, an inverse parameter (γ) may be determined using the mean processing time (μ), the standard deviation (σ), and the confidence level (1-α). Further, a queue length (k) may be computed based upon the total processing time (T), the mean processing time (μ), the standard deviation (σ), and the inverse parameter (γ). Finally, the queue length (k) and the number of jobs (n) may be processed in order to determine the number of computing resources.