Server Pool Management via Capacity Utilization Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing server pool management systems face inefficiencies in energy usage and performance due to underutilization or overutilization of server capacity, leading to increased energy consumption and potential performance penalties, particularly in data centers where temperature control is a concern.

Innovation Solution

A computer-implemented method that monitors system capacity utilization, predicts critical utilization periods, and dynamically adjusts the number of active servers by increasing them when the predicted duration exceeds a threshold, using capacity monitoring components to optimize resource utilization and reduce energy waste.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If servers are added to the active pool to handle increased demand, then system capacity and performance are improved, but energy consumption increases due to start-up and idle power usage

Engineering Contradiction:

Improvesystem capacityVSAvoidenergy consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary analysis by detecting critical utilization conditions and predicting their duration before making server activation decisions. This allows the system to prepare and execute capacity adjustments only when truly necessary, avoiding premature server start-ups that would consume energy without providing proportional performance benefits

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors system capacity utilization and uses this feedback to dynamically adjust the number of active servers. By establishing a feedback loop between utilization monitoring and server management, the system maintains optimal capacity while minimizing energy waste from both over-provisioning and frequent start-stop cycles

Inventive Principle:

Principle #23Feedback

2Use of energy by moving object

If servers are removed from the active pool to reduce energy consumption, then energy efficiency is improved, but system performance deteriorates due to insufficient capacity

Engineering Contradiction:

Improveenergy efficiencyVSAvoidsystem performance

Core Design Contradiction:

Use of energy by moving objectVSProductivity

Solution Approach 1:

The system predicts the duration of critical utilization periods in advance before removing servers from the active pool. By forecasting whether high utilization is temporary or sustained, the system can make informed decisions about server deactivation, ensuring performance requirements are met while maximizing energy efficiency during appropriate periods

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Continuous monitoring of system capacity utilization provides feedback that guides dynamic adjustment of active server count. The system uses this real-time information to balance performance and energy efficiency, activating additional servers only when utilization indicates genuine need rather than temporary fluctuations

Inventive Principle:

Principle #23Feedback

3Productivity

If the number of active servers is frequently adjusted to maintain constant utilization, then resource allocation efficiency is improved, but system stability deteriorates due to start-up and shut-down penalties

Engineering Contradiction:

Improveresource allocation efficiencyVSAvoidsystem stability

Core Design Contradiction:

ProductivityVSStability of the object's composition

Solution Approach 1:

The system detects critical utilization conditions and predicts their duration before making any server activation or deactivation decisions. This preliminary analysis filter prevents frequent, unnecessary adjustments by only triggering capacity changes when the predicted duration of critical utilization exceeds a threshold, thereby maintaining system stability while still achieving resource allocation efficiency

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system cushions against the instability caused by frequent server adjustments by requiring that critical utilization conditions persist for a predicted duration exceeding a defined threshold before triggering capacity changes. This cushioning mechanism filters out transient fluctuations and prevents premature adjustments that would cause start-up and shut-down penalties

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS10171572B2Server pool management

Publication Date: 2019.01.01 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Managing a system comprising a pool of servers including a number of active servers. Utilization of the system capacity is monitored, in which the utilization results from a workload of the number of active servers. Upon detecting a critical utilization of the system capacity during the monitoring of the utilization of the system capacity, a duration of the detected critical utilization of the system capacity is predicted from a monitoring history of the utilization of the system capacity. Further, the number of active servers is increased, if the predicted duration exceeds a defined threshold. Tuning measures may also be deployed to handle surges in capacity utilization.