SciTransfer
OPERA · Project

Energy-Efficient Computing Hardware That Cuts Data Center and Embedded System Power Bills

digitalPilotedTRL 6

Imagine your computer and your data center are like cars stuck in traffic — engines running full blast even when they're barely moving. OPERA built smarter engines that automatically shift gears depending on the workload, using a mix of regular processors and special-purpose chips (like FPGAs) that can be reconfigured on the fly. They tested this on three real scenarios: monitoring road traffic, powering virtual desktops, and running computers inside trucks for civil protection. The result is computing that does the same job while burning far less electricity.

By the numbers
3
Real-life use cases validated (traffic monitoring, truck, virtual desktop)
11
Consortium partners across 5 countries
67
Total project deliverables produced
10
Demonstration and prototype deliverables
7
Industry partners in the consortium
4
SMEs participating in the project
The business problem

What needed solving

Data centers and embedded computing systems waste enormous amounts of electricity running general-purpose processors at full power even when workloads are light or could be handled more efficiently by specialized hardware. Companies running cloud infrastructure, IoT sensor networks, or mobile computing platforms face growing energy bills and tightening EU efficiency mandates — but retrofitting existing systems with energy-smart hardware is technically complex and risky without proven reference designs.

The solution

What was built

OPERA produced working Ultra-Low Power hardware prototypes (multiple releases), FPGA accelerator card prototypes integrated into HP Moonshot servers, an energy efficiency module for cloud resource allocation, workload characterization tools, and energy-performance modeling benchmarks. All of this was validated across 3 demonstration deployments: traffic monitoring in Region Isère, truck-mounted systems for Civil Protection, and scalable virtual desktop data centers.

Audience

Who needs this

Data center operators looking to cut energy costs without sacrificing performanceIoT and edge computing companies needing ultra-low-power field devicesFleet management firms requiring in-vehicle real-time data processingSmart city infrastructure providers deploying roadside sensors and monitoringCloud service providers seeking workload-aware resource allocation to reduce waste
Business applications

Who can put this to work

Cloud Computing & Data Centers
enterprise
Target: Data center operators and cloud service providers

If you are a data center operator dealing with rising electricity costs and cooling demands — this project developed an energy efficiency module that allocates cloud instances to the most appropriate hardware resources based on actual load patterns. They validated workload characterization and resource allocation across 3 real-life use cases with 67 deliverables covering the full stack from hardware to orchestration.

Embedded Systems & IoT
SME
Target: Companies building edge computing or IoT sensor platforms

If you are an IoT platform company struggling with power consumption in field-deployed devices — OPERA designed and prototyped Ultra-Low Power hardware that connects smart sensors to remote small form-factor data centers. The ULP platform went through multiple hardware integration releases and field testing, proving that embedded systems can run on minimal energy while still offloading heavy processing to the cloud.

Smart Transportation & Logistics
mid-size
Target: Fleet management and intelligent transport companies

If you are a transport or logistics company needing real-time data processing in vehicles or roadside infrastructure — OPERA demonstrated a truck-mounted computing system deployed at a Civil Protection test site and a traffic monitoring system tested in Region Isère, France. These use cases proved that low-power heterogeneous computing can handle real-time processing in mobile and outdoor environments.

Frequently asked

Quick answers

What would it cost to adopt this technology?

The project did not publish pricing or licensing fees. Since the coordinator is STMicroelectronics — a major semiconductor manufacturer — any commercial products based on OPERA results would likely be priced as enterprise hardware or IP licensing. Contact the consortium for specific pricing discussions.

Can this scale to large data center deployments?

OPERA specifically targeted scalable architectures. They validated a 'Scalable Small Form Factor Data Center' in the virtual desktop use case and developed resource allocation algorithms that adapt to changing load patterns. The consortium included 11 partners with 7 from industry, suggesting the designs were built with commercial scale in mind.

What about IP and licensing?

OPERA was a Research and Innovation Action under Horizon 2020, meaning IP typically stays with the partners who generated it. STMicroelectronics (coordinator) and HP (contributed to FPGA integration) are well-established companies with existing IP licensing infrastructure. Specific licensing terms would need to be negotiated with the relevant partner.

How mature is the technology — is it ready to deploy?

The project produced working prototypes including ULP hardware platforms (multiple integration releases) and FPGA accelerator cards. These were tested across 3 demonstration use cases at real test sites. However, since the project ended in 2018, further development may have occurred within partner companies.

How does this integrate with existing infrastructure?

OPERA was designed around heterogeneous architectures — meaning it works alongside existing processors by adding FPGA accelerator cards and ULP modules. The FPGA designs were integrated into existing HP Moonshot server platforms, showing compatibility with commercial hardware.

What compliance or regulation does this address?

While OPERA did not target specific regulations, EU energy efficiency directives increasingly require data centers to reduce power consumption. The project's energy-performance modeling and workload-aware resource allocation directly support compliance with energy reporting and reduction mandates.

Is there ongoing support or follow-up?

The project ended in November 2018. Based on available project data, there is no public information about a direct follow-up project. However, STMicroelectronics and the other industry partners may have continued development commercially. The project website operaproject.eu may have further updates.

Consortium

Who built it

The OPERA consortium is heavily industry-driven: 7 out of 11 partners come from industry, giving a 64% industry ratio — well above average for EU research projects. The coordinator, STMicroelectronics, is one of Europe's largest semiconductor companies, which adds serious weight to any hardware IP that came out of this project. With 4 SMEs in the mix and partners spread across 5 countries (France, Israel, Italy, Netherlands, UK), the consortium covers the full value chain from chip design to system integration to end-user deployment. The presence of HP (contributing to FPGA-Moonshot integration) and CSI (handling test site deployments) means the results were tested against real commercial and public-sector requirements, not just lab conditions.

How to reach the team

STMicroelectronics SRL (Italy) — contact their R&D or technology licensing division for OPERA results

Next steps

Talk to the team behind this work.

Want to explore how OPERA's low-power computing technology could reduce your infrastructure energy costs? SciTransfer can connect you with the right consortium partner for your specific use case.