Dynamic Resource Scaling for Distributed Database Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed database services face inefficiencies due to static resource allocation thresholds, leading to underutilization of computational resources during non-peak times, resulting in wasted resources and increased costs.
Innovation Solution
A resource allocator dynamically adjusts computational resource thresholds by predicting utilization through wavelet transformation and autoregressive-moving average modeling of resource consumption data, allowing for real-time adjustment of request units based on predicted values and standard deviation buffers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static resource allocation thresholds are used, then resource allocation is simple to manage, but resource underutilization occurs during non-peak times leading to waste
Solution Approach 1:
The patent applies dynamics by transitioning from static resource allocation thresholds to dynamic thresholds that automatically adjust based on real-time resource consumption patterns. The system continuously monitors resource usage and updates allocation thresholds to match actual demand, ensuring resources are neither over-provisioned nor under-provisioned. This dynamic adaptation eliminates resource waste during low-utilization periods while maintaining adequate capacity during peak periods.
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring resource consumption data and using this information to adjust allocation thresholds. The system collects historical resource usage data, analyzes consumption patterns, and feeds this information back into the resource allocation decision-making process. This closed-loop feedback system enables automatic optimization of resource allocation without manual intervention, resolving the contradiction between ease of management and resource efficiency.
2Reliability
If allocated threshold is increased to combat peak usage, then resource availability during peak times is improved, but resource under-utilization occurs during non-peak times
Solution Approach 1:
The system uses dynamic threshold adjustment to make resource allocation flexible rather than fixed. Allocation thresholds automatically scale up during peak usage periods to ensure adequate resource availability and scale down during non-peak periods to maximize utilization efficiency. This dynamic behavior allows the system to maintain high reliability during critical periods while optimizing productivity during lower-demand periods.
Solution Approach 2:
The patent changes the parameter of allocation threshold from a fixed value to a variable that adapts based on resource consumption patterns. By continuously adjusting the threshold parameter in response to changing workload conditions, the system achieves both high availability during peaks and high efficiency during non-peaks, resolving the contradiction between reliability and productivity.
3Stability of the object's composition
If static allocation threshold is used, then resource allocation is predictable and stable, but resource waste occurs due to inability to adapt to changing demand
Solution Approach 1:
The patent introduces dynamic adaptation while maintaining stability through systematic analysis of resource consumption patterns. Rather than making arbitrary changes, the system uses structured methods to adjust allocation thresholds based on observed usage trends. This approach preserves the stability and predictability of resource allocation while enabling adaptation to changing demand, thereby eliminating resource waste without sacrificing system stability.
Data Source
AI summary
Described embodiments provide systems and methods for resource scaling. A computing device may include a data processor. The data processor may receive, from a distributed database service (DDS), usage data on resource request-units from a plurality of services that use resources provided by the DDS. The computing device may include a transform agent. The transform agent may transform at least a subset of the N sets of data, into at least three transformed datasets each corresponding to a different frequency range. The computing device may include a predictor. The predictor may provide a predicted value of request-units according to the at least three transformed datasets. The predictor may request the DDS to provide the resources up to a number of request-units at a given time instance according to the predicted value.