RAVEL and Sandia National Laboratories Explore Power-Aware AI Infrastructure for a More Efficient, Competitive AI Future

Project Quicksilver uses RAVEL Orchestrate AI to investigate distributed workload management, energy efficiency, infrastructure economics, and performance

SEATTLE, Nov. 12, 2026 /PRNewswire/ -- RAVEL today announced a research collaboration with Sandia National Laboratories on Project Quicksilver, a multi-phase initiative using RAVEL Orchestrate AI to investigate dynamic, power-aware orchestration of artificial intelligence (AI) workloads across geographically distributed computing environments. The research aims to inform more efficient, responsible, and cost-effective approaches to data center design and operations by examining how power availability, operational conditions, and workload requirements can guide scheduling and workload placement decisions to address energy, capacity, and cost constraints.

The race to expand AI infrastructure is creating a fundamental challenge: more computing capacity does not automatically translate into more productive capacity. As power availability, infrastructure costs, regulatory requirements, and workload demands increasingly shape AI infrastructure decisions, Project Quicksilver investigates how dynamic, power-aware orchestration can help organizations make more productive use of power and compute capacity across distributed environments.

"The next phase of AI infrastructure leadership won't be won by simply building more. It will be determined by how efficiently and productively we use every megawatt, every GPU, and every dollar invested," said Denise Muyco, CEO and Co-founder of RAVEL. "The United States has an opportunity to continue to lead and innovate not just in AI capacity, but in how intelligently that capacity is managed. Project Quicksilver will investigate how orchestration can help make AI infrastructure investments more productive while supporting more responsible energy use."

Investigating dynamic, power-aware orchestration

Project Quicksilver examines how AI workloads can be scheduled, balanced, or moved across distributed computing environments in response to changing power availability and operating conditions. Using Box–Behnken and Plackett–Burman experimental design methodologies respectively, Phase One will evaluate air-cooled and immersion-cooled computing environments across three Sandia locations connected by a wide-area network. The research will vary network bandwidth and speed alongside other experimental factors to investigate three key areas:

• Energy efficiency: Evaluate the energy consumption and operating costs of identical workloads across cooling environments, accounting for compute, IT infrastructure, and data center operations, including energy use, TUE and PUE-adjusted efficiency, and potential heat-reuse benefits.
• Total cost of ownership (TCO): Evaluate how workload scheduling and placement decisions affect computing costs across distributed environments, including capital expenditures (CAPEX) and operating expenditures (OPEX) associated with infrastructure, power, cooling, networking, and operations.
• Workload time-to-completion: Evaluate how workload scheduling and placement decisions impact AI workload execution time.

"Many laboratories and data centers around the world look to Sandia's work in developing and deploying energy-efficient designs and water-saving technologies," said Dave Martinez, Engineering Program and Project Lead for Infrastructure Computing Services at Sandia National Laboratories. "Managing workloads across multiple computing environments creates opportunities, while also introducing complex trade-offs. Under a Cooperative Research and Development Agreement with RAVEL, we'll explore whether dynamic workload placement can make better use of available power and computing resources while balancing performance and cost."

Federated, power-aware orchestration with RAVEL Orchestrate AI

RAVEL Orchestrate AI provides a federated orchestration layer for managing AI workloads across heterogeneous, geographically distributed computing resources. Rather than optimizing each environment in isolation, federation enables workloads to be managed across locations as part of a coordinated system, with scheduling and placement decisions informed by available power, cost, and performance requirements. The platform combines power-aware scheduling and policy-driven management to account for infrastructure conditions, workload requirements, and operational priorities.

"AI infrastructure can't reach its full potential if every data center operates as an island," said Philippa Carroll, Chief Product Officer at RAVEL. "As AI infrastructure becomes more distributed, the challenge is no longer just how much computing capacity we have, but how effectively we can use it. Project Quicksilver will investigate the trade-offs between power, cost and performance—and the role orchestration plays in ensuring we make better and more responsible infrastructure decisions."

Informing future AI infrastructure design and operations

Project Quicksilver aims to generate evidence that could inform future AI infrastructure design and operations, including approaches to managing workloads across distributed computing environments.

Sandia serves as the initial research lead, with plans to expand Project Quicksilver to additional U.S. Department of Energy (DOE) national laboratories and explore additional use cases for environmental and performance optimization.

Learn more at SC26

RAVEL and Sandia will discuss Project Quicksilver at SC26, November 15–20, 2026.

Project Quicksilver: Dynamic Power-Aware AI Orchestration Across Distributed Computing Environments

Tuesday, November 17, 2026 | 2:30 p.m. CST
U.S. Department of Energy Booth | SC26

The presentation will cover Project Quicksilver's experimental approach, planned test phases, and anticipated research outputs.

Visit the DOE booth at SC26 to learn more about Project Quicksilver. 

About RAVEL

RAVEL is redefining intelligent orchestration for IT operations. Through its Orchestrate AI platform, RAVEL empowers teams to scale innovation, reduce infrastructure friction, and unlock the full potential of AI—from research to production.

SOURCE RAVEL Inc.