20 Ways Engineering Teams Can Optimize Workloads For Energy Efficiency

Energy efficiency is a growing priority in the engineering space, and teams must find creative ways to optimize workloads without compromising performance. Small changes, like redistributing work in real time or leveraging cloud-based solutions, can lead to significant reductions in energy consumption.

To help you make your team more energy efficient, Forbes Technology Council members weigh in with effective approaches to reduce energy consumption and maximize performance. These expert insights can help improve team sustainability without sacrificing speed or functionality.

1. Implement Dynamic Workload Orchestration

Dynamic workload orchestration improves energy efficiency by distributing workloads based on real-time demand, hardware efficiency and power availability. Using AI-driven scaling, resource pooling and load balancing, teams can minimize idle compute power, reduce energy waste and maintain peak performance without compromising system reliability or responsiveness. – Nicola Sfondrini, PWC

2. Enable Cloud Operations Technology

I’m seeing more organizations investigating the NVMe over TCP storage protocol, which is an enabler of cloud operations models. It provides high performance and consistently low latency at scale for performance-intensive workloads while reducing hardware and heat transfer in data centers, which ultimately improves energy efficiency. – Abel Gordon, Lightbits Labs

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?

3. Apply Intelligent Workload Scheduling

One practical way is to employ intelligent workload scheduling and auto-scaling. Cloud-native solutions allow teams to scale compute resources up or down based on demand, ensuring optimal utilization and energy efficiency. Implementing efficient coding practices and serverless architectures can also lead to significant energy savings. – Preetpal Singh, Xebia

4. Develop An Energy Management System

The goal is to develop a dynamic Energy Management System (EMS). In industry, EMS or BEMS collect energy consumption data, which, when analyzed with machine learning algorithms, optimizes process performance and efficiency. Additionally, this data supports predictive maintenance, preventing inefficiencies and ensuring seamless operations. – Ilker Kalali, Pirelli Tire North America

5. Break Down Projects Into Small Components

Focus on simplicity and optimization. Engineering teams can leverage modular design, breaking down projects into smaller, more efficient components. This minimizes wasted resources and streamlines workflows, allowing for high performance without excess. Less is more—it’s about doing more with less, which leads to both energy and resource efficiency. – Oleg Sadikov, DeviQA

6. Use Containerization And Automation

Engineering teams can optimize workloads by using containerization, like Docker, to package applications for consistent deployment and orchestration, such as Kubernetes, to automate scaling and management. This approach enhances resource utilization, speeds up deployments and increases reliability through automated workload distribution. – Ambika Saklani Bhardwaj, Walmart Inc.

7. Implement A Resource Monitoring And Optimization Framework

To maximize energy efficiency, companies must strategize the implementation of a resource monitoring and optimization framework, which their engineering teams can then adopt. This framework can provide AI-driven automated recommendations, such as right-sizing instances, standardized configurations and continuous feedback to ensure consistent and effective energy savings. – Sibin Thomas, Google

8. Optimize Storage And Scaling With AWS

Our services run on AWS, optimizing workloads through dynamic resource allocation. AWS Auto Scaling ensures EC2 instances and RDS databases are rightsized, preventing over-provisioning and improving efficiency. For storage, we use S3 Intelligent-Tiering to automatically shift infrequently accessed data to lower-energy storage classes, reducing energy consumption without impacting performance. – Jason Penkethman, Simpro Group

9. Utilize GenAI For Quality Checks And Support

Engineering teams can reduce workloads by using GenAI to automate time-consuming tasks like quality checks and development support. By automating these processes, you can supercharge your employees so they save time, use resources more efficiently and keep performance high without sacrificing the quality of their work or burning out. – Adam Lieberman, Finastra

10. Understand Your Technology’s Purpose To Identify Key Metrics

Engineers leverage observability but often struggle with efficiency. A “collect everything” approach creates a data deluge that slows tools, fuels false alarms and extends mean time to repair. Instead, start with your technology’s purpose—why it exists—and drill down logically to identify the right metrics. A targeted approach delivers better insights, reduces noise and improves performance. – Bill Hineline, Chronosphere

11. Automate Metadata-Driven Data Tiering

Metadata analysis can identify and automate data placement across storage tiers, reducing redundant storage and aligning resources with data value. This approach minimizes energy-intensive over-provisioning by archiving cold or obsolete data to cost-efficient, low-power tiers while ensuring access to critical data without sacrificing performance. – Carl D’Halluin, Datadobi

12. Optimize Network Traffic

Start by optimizing traffic to your network, security and application tools in your hybrid multi-cloud infrastructure. By doing so, and increasing visibility into all network traffic, you can significantly reduce traffic volumes. This then reduces power consumption massively, improves efficiency, supports your bottom line and improves your carbon footprint. – Shane Buckley, Gigamon

13. Eliminate Redundant Computations

Optimizing software to eliminate redundant computations and using energy-efficient hardware, like GPUs for parallel tasks, significantly enhances performance per watt. This reduces power consumption without compromising output. By aligning software and hardware optimization, systems can achieve high performance while minimizing energy use, ensuring both efficiency and sustainability. – Dhivya Nagasubramanian, U.S. Bank

14. Switch Long-Running Containers To Serverless Functions

Switching from long-running containers to event-based serverless functions can not only improve energy efficiency but also give a speed boost while cutting costs. Today’s WebAssembly-based serverless functions are much faster than AWS Lambda and the first-gen tooling. – Matt Butcher, Fermyon Technologies

15. Consolidate Data Into Fewer Streamlined Workflows

Consolidate fragmented data operations into fewer, more streamlined workflows, reducing context switching, improving cache efficiency and lowering compute overhead with more orchestration. Not only does this reduce redundant processing, but it also gives you more visibility into performance bottlenecks so you can optimize them in detail. – Sandro Shubladze, Datamam

16. Implement Auto-Scaling With Dynamic Thresholds

Implement machine-learning-driven auto-scaling with dynamic thresholds instead of static rules. By predicting workload patterns, teams can proactively adjust resources before they’re needed while maintaining performance. This precise resource allocation reduces compute usage, cooling requirements and costs—all without sacrificing business outcomes. – Kim Bozzella, Protiviti

17. Forecast Resource Needs And Allocate Tasks

Implementing proactive workload scheduling with AI-driven predictive analytics is key. By forecasting resource needs and dynamically allocating tasks to optimal time slots and hardware, engineering teams can minimize energy consumption during low-demand periods, maintaining top performance while significantly improving energy efficiency and reducing costs. – Aravind Nuthalapati, Microsoft

18. Find Low Carbon Footprint Regions On Google Cloud

On Google Cloud, the carbon footprint of individual regions (each of which has unique power utilization efficiency and upstream power sources) is made easy to find with a little green leaf and a “Low CO2” highlight. Pick those regions, and do everything the same, and you’re making a positive difference. – Miles Ward, SADA, An Insight company

19. Use Cloud-Native Tools To Analyze Configurations And Metrics

Engineering teams can use cloud-native tools like AWS Compute Optimizer to analyze resource configurations and utilization metrics. It provides rightsizing recommendations, balancing cost and performance while ensuring capacity needs. By leveraging insights from historical and projected usage data, teams can strategically adjust workloads to enhance efficiency without sacrificing performance. – John Anand Lourdusamy, Capital One

20. Find A Carbon-Aware Workload Scheduler

One overlooked way to optimize workloads for energy efficiency is to adopt GreenOps, a cloud sustainability practice that integrates real-time carbon-aware workload scheduling. By dynamically shifting non-urgent tasks to off-peak hours or renewable-powered regions, engineering teams can cut emissions, lower energy costs and improve sustainability without sacrificing performance. – Jabin Geevarghese George, Tata Consultancy Services