Self-adaptive server capacity control for latency and energy trade-offs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing load balancing methods in server clusters do not effectively address power and energy conservation, as they rely on fixed server provisioning that leads to inefficient power usage due to servers operating at low CPU utilization, and existing dynamic capacity management techniques are inadequate for dynamically evolving web applications.

Innovation Solution

A self-adaptive control system with a centralized controller that uses real-time and historical data to dynamically adjust the number of active servers based on workload changes, employing classic control theory to optimize server capacity and reduce latency while maximizing energy savings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If fixed server provisioning is used to ensure adequate capacity for latency-sensitive applications, then response time performance is maintained, but power consumption increases due to servers operating at low CPU utilization

Engineering Contradiction:

Improveresponse timeVSAvoidpower consumption

Core Design Contradiction:

Loss of timeVSUse of energy by stationary object

Solution Approach 1:

The patent implements dynamic capacity management that automatically adjusts the number of active servers based on real-time workload conditions. The system transitions from fixed provisioning to dynamic scaling, where server capacity is adjusted continuously to match actual demand, thereby maintaining performance during peak loads while reducing power consumption during low-utilization periods

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system employs feedback control mechanisms that monitor CPU utilization metrics and use this information to make informed decisions about server activation and deactivation. The feedback loop ensures that capacity adjustments are made based on actual system state, preventing both performance degradation and unnecessary power consumption

Inventive Principle:

Principle #23Feedback

2Use of energy by stationary object

If the number of active servers is reduced to save energy, then power consumption decreases, but system adaptability to dynamically evolving web applications deteriorates

Engineering Contradiction:

Improvepower consumptionVSAvoidsystem adaptability

Core Design Contradiction:

Use of energy by stationary objectVSAdaptability or versatility

Solution Approach 1:

The system implements dynamic capacity management that automatically adjusts the number of active servers based on real-time workload conditions. The system transitions from fixed provisioning to dynamic scaling, where server capacity is adjusted continuously to match actual demand, thereby maintaining performance during peak loads while reducing power consumption during low-utilization periods

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The capacity management system operates autonomously, using built-in monitoring and control mechanisms to self-adjust server capacity without external intervention. The system monitors its own performance metrics and automatically makes capacity decisions, enabling it to adapt to changing workload patterns while optimizing energy consumption

Inventive Principle:

Principle #25Self-service

3Reliability

If more servers are kept active to handle peak workloads, then system reliability is improved, but energy waste increases due to servers operating at low utilization

Engineering Contradiction:

Improvesystem reliabilityVSAvoidenergy waste

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the operational parameters of server capacity by adjusting the number of active servers based on workload conditions. This parameter adjustment allows the system to maintain reliability during high-demand periods while minimizing energy waste during low-utilization periods through controlled parameter changes

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10212220B2Self-adaptive control system for dynamic capacity management of latency-sensitive application servers

Publication Date: 2019.02.19 META PLATFORMS INC

AI summary

A self-adaptive control system based on proportional-integral (PI) control theory for dynamic capacity management of latency-sensitive application servers (e.g., application servers associated with a social networking application) are disclosed. A centralized controller of the system can adapt to changes in request rates, changes in application and/or system behaviors, underlying hardware upgrades, etc., by scaling the capacity of a cluster up or down so that just the right amount of capacity is maintained at any time. The centralized controller uses information relating to a current state of the cluster and historical information relating to past state of the cluster to predict a future state of the cluster and use that prediction to determine whether to scale up or scale down the current capacity to reduce latency and maximize energy savings. A load balancing system can then distribute traffic among the servers in the cluster using any load balancing methods.