Capacity Manager for Batch Workload Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face challenges in efficiently managing computing resources to match available types with required workloads, particularly in ensuring time-critical transactional workloads are completed on time without utilizing priority capacity for less time-sensitive batch workloads.
Innovation Solution
The system employs a capacity manager that monitors processor consumption, calculates a consumption ratio, and allocates workloads to control the use of priority capacity by deferring or delaying batch processing when necessary, ensuring priority capacity is reserved for peak workloads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If batch workloads are processed using available computing resources, then productivity is improved, but priority capacity may be consumed by non-time-critical jobs
Solution Approach 1:
The system continuously monitors processor consumption and dynamically adjusts batch workload processing based on real-time capacity usage. The capacity manager receives feedback about current processor consumption levels and automatically modulates batch job execution to prevent priority capacity depletion, ensuring time-critical workloads always have resources available.
Solution Approach 2:
The system implements dynamic workload management where batch processing intensity is adjusted in real-time based on current system conditions. When priority capacity is being consumed, batch workloads are automatically throttled or deferred. When capacity is available, batch processing intensifies. This dynamic adjustment resolves the contradiction by making batch processing flexible rather than static.
2Adaptability or versatility
If computing resources are allocated to meet diverse workload requirements, then adaptability is improved, but device complexity increases
Solution Approach 1:
The system segments computing resources into distinct categories: base capacity for batch workloads and priority capacity for time-critical workloads. This segmentation allows the system to manage different workload types independently through a capacity manager that monitors and controls resource allocation separately for each capacity type, reducing overall management complexity while maintaining adaptability.
Solution Approach 2:
The capacity manager acts as an intermediary layer between the physical computing resources and the diverse workloads. It translates various workload requirements into appropriate resource allocations by monitoring processor consumption and automatically adjusting batch workload execution. This intermediary simplifies the complexity by providing a single point of control that handles all resource allocation decisions.
Data Source
AI summary
Described techniques obtain processor consumption within a first time interval by a subset of processors of a plurality of processors of at least one computing device to update a rolling average of processor consumption of the subset of processors over a second time interval. By converting the rolling average to a consumption rate and determining a consumption ratio of the consumption rate to a base capacity provided by the subset of processors, described techniques may allocate a workload using the subset of processors, and based on the consumption ratio, to thereby control use of a priority capacity in processing the workload.