Cloud DBMS Workload Selection for Cost and Deadline Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in accurately estimating runtime, capacity, and query price in real-time, especially in cloud environments with dynamic resource allocation and cost-sensitive operations.
Innovation Solution
The proposed solution involves a database management system (DBMS) that determines multiple workloads qualified for executing a request based on cost and elapsed time criteria, allowing for self-tuning and adaptive capabilities to optimize workload management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cloud providers provision infrastructure to users' peak demands, then service level goals are met, but resource utilization efficiency deteriorates with idle resources at off-peak times
Solution Approach 1:
The patent implements dynamic workload management that automatically adjusts resource allocation based on real-time conditions. The system transitions between different workload states (e.g., from peak to off-peak) and dynamically provisions or de-provisions resources accordingly, allowing the infrastructure to adapt to changing demand patterns while maintaining service level goals.
Solution Approach 2:
The system performs preliminary actions by pre-warming or pre-provisioning resources before peak demand occurs, and by gracefully de-provisioning resources during off-peak periods. This allows the system to meet peak demands without over-provisioning for the entire period, thereby reducing idle resource waste.
2Stability of the object's composition
If volume-discounted pricing is offered to incentivize larger, longer-lived jobs, then resource demand stability improves, but flexibility to handle variable workloads deteriorates
Solution Approach 1:
The patent segments workloads into different categories or tiers based on their characteristics (e.g., predictable vs. variable, batch vs. interactive). This segmentation allows the system to apply different pricing and resource allocation strategies to different workload types, thereby maintaining demand stability for predictable workloads while preserving flexibility for variable workloads.
Solution Approach 2:
Different pricing policies and resource allocation rules are applied locally to different workload types rather than uniformly across all workloads. Volume discounts and committed use discounts are applied specifically to workload categories that benefit from stability, while spot pricing or on-demand pricing remains available for workloads requiring flexibility.
3Loss of energy
If spot instances are used to reduce cloud costs, then cost efficiency improves, but job interruption risk increases due to capacity reclamation
Solution Approach 1:
The system implements beforehand cushioning by creating redundancy or backup capacity for critical workloads running on spot instances. When spot instance capacity is reclaimed, the system has pre-established mechanisms to migrate or re-provision the workload, thereby cushioning against job interruptions and maintaining reliability despite cost optimization efforts.
Solution Approach 2:
The system employs feedback mechanisms that continuously monitor spot instance availability, workload performance, and cost metrics. Based on this feedback, the system dynamically adjusts workload placement decisions, transitions between instance types, or triggers migration events to maintain job continuity while optimizing costs.
Data Source
AI summary
A computer system executes a database management system (DBMS). The DBMS manages a database comprised of DBMS resources. The DBMS receives a request to be executed. The request is a DBMS action to be executed using the DBMS resources. The request includes a predicate specifying a maximum cost for executing the request, and a deadline, specifying a deadline by which the request is to be completed in its execution. The DBMS determines a plurality of workloads under which the request is qualified to execute. Each workload of the plurality of workloads includes a respective set of requests that have common characteristics. Each workload of the plurality of workloads includes a respective cost criterion and a respective elapsed time criterion. The DBMS selects a selected workload from among the plurality of workloads. The selected workload has a selected cost criterion and a selected elapsed time criterion. The DBMS begins execution of the request using the selected workload.