Multicore Processor Workload Scheduling for Server Capacity Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computing systems face inefficiencies as they operate multiple servers at low capacity, leading to increased computing costs and space requirements, necessitating a more efficient strategy to maximize processor and server utilization.
Innovation Solution
A method and system that schedules computational tasks to maximize the average number of processing cores utilized per clock cycle by distributing tasks among multicore processors, using machine learning to classify and group similar tasks, and optimizing workload schedules to ensure efficient RAM utilization across processors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computational workloads are distributed over many servers, then system reliability and task completion are improved, but computing costs and space requirements increase
Solution Approach 1:
The patent consolidates computational workloads onto fewer servers by maximizing processor utilization through intelligent task scheduling. Multiple tasks are merged and executed on the same server resources, reducing the total number of servers needed while maintaining system reliability through efficient resource sharing and load management.
Solution Approach 2:
The patent makes server resources universal by enabling a single server to handle multiple different computational tasks through dynamic scheduling. The same physical server infrastructure is used to execute diverse workloads at different times, maximizing the utility of each server and reducing the total quantity needed.
2Ease of operation
If servers are operated at low capacity, then task distribution is simplified, but computing costs and space requirements increase
Solution Approach 1:
The patent introduces dynamic task scheduling that adapts server capacity allocation based on workload demands. Instead of static low-capacity operation, the system dynamically adjusts the number and capacity of active servers, optimizing resource utilization while maintaining operational simplicity through automated scheduling algorithms.
Solution Approach 2:
The patent changes the operational parameter of server capacity from fixed low utilization to variable high utilization. By modifying how servers are operated (from underutilized to fully utilized through scheduling), the system reduces the total number of servers required while keeping task distribution manageable through systematic scheduling.
3Productivity
If more processors are used, then computational throughput is improved, but device complexity and management difficulty increase
Solution Approach 1:
The patent segments computational tasks into discrete units that can be scheduled and managed individually. This segmentation allows the system to handle complex processor workloads by breaking them down into manageable task components, maintaining high productivity while reducing the complexity of processor management through structured task decomposition.
Solution Approach 2:
The patent implements feedback mechanisms through task scheduling algorithms that monitor processor utilization and adjust task allocation accordingly. This feedback loop enables the system to maintain high computational throughput while automatically managing processor complexity, as the scheduling system responds to actual processor states and optimizes resource allocation dynamically.
Data Source
AI summary
A system and method for operating fewer servers near maximum capacity as opposed to operating more servers at low capacity is disclosed. Computational tasks are made as small as possible to be completed within the available capacity of the servers. Computational tasks that are similar may be distributed to the same computing node (including a processor) to improve RAM utilization. Additionally, workloads may be scheduled onto multicore processors to maximize the average number of processing cores utilized per clock cycle.