Automated Workload Placement for Data Center Consolidation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Data centers face increasing complexity and costs due to the exponential growth of workload combinations as the number of servers and workloads increases, leading to under-utilization of resources and inefficient allocation of computing services.

Innovation Solution

An automated workload placement system that uses historical data and constraints to recommend optimal placement strategies, such as consolidation and load balancing, to efficiently allocate workloads across computing resources, reducing the burden on administrators and minimizing energy and maintenance costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the number of servers is increased to accommodate more workloads, then the computing service capacity is improved, but the complexity of workload allocation increases exponentially

Engineering Contradiction:

Improvecomputing service capacityVSAvoidworkload allocation complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces an automated workload placement system that acts as an intermediary between administrators and the complex server infrastructure. This system uses historical data and constraints to automatically determine optimal workload placements, eliminating the need for administrators to manually manage the exponentially increasing allocation complexity while maintaining high computing service capacity

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system enables self-service by automatically performing workload placement decisions without requiring administrator intervention. The automated system analyzes historical data, evaluates constraints, and makes placement decisions independently, allowing the infrastructure to manage itself and freeing administrators from complex allocation tasks

Inventive Principle:

Principle #25Self-service

2Productivity

If the number of servers is increased to handle more workloads, then the computing capacity is improved, but the power and cooling costs increase

Engineering Contradiction:

Improvecomputing capacityVSAvoidpower and cooling costs

Core Design Contradiction:

ProductivityVSUse of energy by stationary object

Solution Approach 1:

The patent applies consolidation strategies that merge multiple workloads onto fewer servers when possible. By analyzing workload characteristics and historical data, the system identifies opportunities to co-locate compatible workloads on the same physical server, thereby reducing the total number of active servers and their associated power and cooling requirements while maintaining adequate computing capacity

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system dynamically changes operational parameters by adjusting workload placement based on real-time and historical performance data. This allows the infrastructure to optimize resource utilization rates, ensuring servers operate at efficient load levels that minimize energy consumption per unit of computing output, thereby reducing overall power and cooling costs

Inventive Principle:

Principle #35Parameter changes

3Productivity

If the number of servers is increased to accommodate more workloads, then the service coverage is improved, but the square footage of the data center increases

Engineering Contradiction:

Improveservice coverageVSAvoiddata center square footage

Core Design Contradiction:

ProductivityVSArea of stationary object

Solution Approach 1:

The patent implements consolidation that merges multiple workloads onto fewer physical servers. By strategically placing workloads on shared infrastructure and eliminating underutilized servers, the system maintains comprehensive service coverage across multiple applications while reducing the physical server count, thereby decreasing the data center square footage required

Inventive Principle:

Principle #5Merging (Combining)

4Reliability

If manual workload allocation is performed, then the control and oversight are maintained, but the time and effort required increases with the number of servers

Engineering Contradiction:

Improveadministrative controlVSAvoidcapacity planning time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system implements feedback mechanisms that continuously monitor workload performance and placement effectiveness. By analyzing historical data and outcomes, the system learns from past decisions and automatically adjusts future placement strategies, maintaining reliable administrative control through configurable constraints while eliminating time-consuming manual iteration

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS8392928B1Automated workload placement recommendations for a data center

Publication Date: 2013.03.05 HEWLETT PACKARD ENTERPRISE DEV LP

AI summary

Automatically recommending workload placement in a data center. Historical data and workload constraints of a plurality of workloads are received. A plurality of resources of the data center available for placement of the plurality of workloads is identified. Resource constraints of the plurality of resources are received. Headroom ratings are assigned to the plurality of workloads based on the historical data, the workload constraints and the resource constraints. A consolidation recommendation of the plurality of workloads on the plurality of the resources in said the center is automatically generated based on an analysis of the historical data, the plurality of workloads, the plurality of available resources in the data center and the headroom ratings.