Compute Environment Management System for Scheduled Reserved Instances
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current managed compute environments lack efficient automated management of computing resources, leading to inefficiencies in resource allocation and utilization, particularly in distributed systems and cloud computing setups, where resources are often underutilized or overutilized due to manual provisioning and lack of dynamic scaling.
Innovation Solution
A compute environment management system that allows clients to define constraints for computing resources, automatically selects and reserves resources based on job requirements and constraints, and dynamically provisions or deprovisions resources to match demand, using scheduled reserved compute instances for predictable workloads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If manual provisioning of computing resources is used, then resource allocation can be controlled, but resource utilization efficiency deteriorates due to underutilization or overutilization
Solution Approach 1:
The system enables self-service through automated resource provisioning and management. The compute environment management system automatically selects, reserves, and provisions computing resources based on job requirements and defined constraints, eliminating the need for manual intervention while optimizing resource allocation and utilization efficiency
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring job requirements, resource availability, and utilization metrics. Based on this feedback, the management system dynamically adjusts resource provisioning, scales resources up or down to match demand, and optimizes the mix of on-demand and scheduled reserved instances to improve overall resource utilization
2Adaptability or versatility
If static resource allocation is used, then resource availability is guaranteed, but adaptability to changing demand deteriorates
Solution Approach 1:
The system applies dynamics by transitioning from static resource allocation to dynamic scaling. The management system continuously adjusts the number and type of computing resources based on real-time job requirements and demand patterns, while maintaining service level agreements through intelligent resource provisioning and orchestration
Solution Approach 2:
The system performs preliminary action by pre-reserving scheduled reserved compute instances based on predictable workload patterns and constraints. This advance preparation ensures resource availability when needed while allowing dynamic adjustment of the reserved capacity to adapt to changing demand over time
3Loss of energy
If on-demand compute instances are used exclusively, then resource availability is high, but cost efficiency deteriorates due to paying for unused capacity
Solution Approach 1:
The system changes parameters by optimizing the mix and allocation of different compute instance types and provisioning models. It dynamically adjusts the proportion of on-demand versus scheduled reserved instances, selects appropriate instance sizes and configurations, and reallocates resources across different job queues to improve cost efficiency while maintaining adequate job execution throughput
Solution Approach 2:
The system applies partial action by provisioning only the necessary amount of computing resources required for current job demands, rather than over-provisioning. It uses scheduled reserved instances for predictable baseline workloads and supplements with on-demand instances only when additional capacity is needed, avoiding payment for excessive unused capacity
4Adaptability or versatility
If scheduled reserved compute instances are used, then cost efficiency improves, but flexibility to handle unpredictable workloads deteriorates
Solution Approach 1:
The system achieves universality by creating a hybrid resource pool that combines scheduled reserved compute instances with on-demand instances. This multi-functional approach allows the same infrastructure to handle both predictable scheduled workloads cost-efficiently and unpredictable ad-hoc workloads flexibly, with the management system intelligently routing jobs to appropriate resource types
Data Source
AI summary
Methods, systems, and computer-readable media for job execution with scheduled reserved compute instances are disclosed. One or more queues are mapped to a compute environment. The queue(s) are configured to store data indicative of jobs. The compute environment is associated with one or more scheduled reserved compute instances, and the one or more scheduled reserved compute instances are reserved for use in the compute environment for a window of time. The queue(s) are mapped to the compute environment prior to the window of time opening. During the window, at least one of the scheduled reserved compute instances is added to the compute environment. The scheduled reserved compute instance(s) are provisioned from a pool of available compute instances. During the window, execution is initiated of one or more jobs from the queue(s) on the scheduled reserved compute instance(s) in the compute environment.