April 20, 2024

Researchers calculate the path to greener computing | News

NREL researchers lead effort to reduce computing emissions


A person presenting behind a podium before a small audience watching.
Charles Tripp, NREL senior researcher and leader of the Green Computing Catalyzer, presents NREL’s green computing efforts at the 2023 JISEA Annual Meeting. Photo by NREL

While it has enabled groundbreaking technological advances, the global growth of advanced computing has created a looming energy crisis. Continuous innovation and increased efficiency are essential to avoid future explosive growth in the energy used for computing; However, there is little established research on how we can better design, manufacture, use and dispose of computing devices to reduce their environmental impact.

The U.S. Department of Energy’s National Renewable Energy Laboratory (NREL) has become a global leader in green computing, both by making its own computing operations more sustainable and by leading research projects to provide new information. and responsibility to computer-based research worldwide. . As part of this effort, NREL’s Joint Institute for Strategic Energy Analysis (JISEA) launched the Green Computing Catalyst to cultivate and scale the lab’s green computing capabilities and establish it as a new research domain.

Long-standing commitment to IT efficiency

As home to two high-performance computing (HPC) systems, Eagle and Kestrel, NREL uses advanced computing to power many of its next-generation modeling, simulation and calculation capabilities. Recognizing the significant energy consumption of these systems, NREL took steps to make their operation more sustainable. In 2014, NREL implemented a water-based cooling system that absorbs waste heat from HPC systems and uses it to heat campus facilities. The researchers also aligned computational jobs with the timing of renewable energy generation, scheduling jobs at optimal times.

In addition to incorporating energy efficiency at the facility and operations level, NREL also implemented a monitoring system to better understand the computational energy consumption of HPC systems. Eagle, Kestrel, and the previous supercomputer, Peregrine, were equipped with sensors to measure the power consumption of individual nodes and components, such as processors, memory, and graphics processing units. This monitoring system helps quantify computational energy consumption in a way that allows for comparison and improvement, and NREL is now working to incorporate these insights into its computational research and practices.

Facilitate energy-efficient computing for HPC users

Researchers at NREL’s Computational Science Center are partnering with the JISEA Green Computing Catalyst to make energy consumption data collected from the supercomputer’s sensors available to all NREL researchers.

“We want to encourage researchers to include data on computational energy consumption as part of their publications,” said Charles Tripp, principal investigator at the Center for Computational Sciences and leader of the JISEA Green Computing Catalyzer. “This draws attention to the energy cost of computing and sets a standard for accountability in this field of research.”

When setting up computational experiments, researchers often compare the performance metrics of various models and algorithms to choose the configuration that will produce the best quality of results. By making algorithmic information on energy consumption accessible, researchers will be able to compare the environmental impact of various computing approaches. The ultimate goal is for any researcher using NREL’s computing facilities to understand the energy consumption of their calculations and to report some degree of energy data in their publications.

“NREL plans to deploy this resource across the HPC user community, so that any researcher using our supercomputers can have the information needed to choose not only the best raw performance but also the optimal power efficiency,” said Aaron Anderson, manager of NREL group. Advanced Computer Operations Group. “Over time, we plan to include more detailed information to make these energy efficiency considerations more useful.”

Leading green computing efforts outside of NREL

In addition to making its own computing more energy efficient, NREL is working to equip researchers around the world with the tools needed to design more energy-efficient algorithms.

Researchers at NREL’s Computational Science Center recently compiled a public dataset of real-world algorithmic performance metrics and ran millions of computational experiments on three HPC systems, reaching a rate of 4 million node hours per year. This substantial amount of computing provided data on the performance of various algorithm configurations, input data, and hardware. As part of the work with the JISEA Green Computing Catalyzer, the researchers are now working to augment the dataset with energy efficiency data. The dataset, called the BUTTER Deep Learning Empirical Dataset, is the first of its kind.

“There are very few computer science publications that publish the power consumption of their approaches on real systems,” Tripp said. “While most publications provide estimates, publishing real-world data is a new frontier.”

Making the data available can help researchers inside and outside of NREL determine how to spend their computational energy wisely and what energy costs they might incur when training models. The researchers are also working to compile and publish similar data sets for cryptocurrency and blockchain calculations.

A plan for the future of green computing research

As green computing research gains greater attention and efforts expand, NREL plans to diversify the types of systems it measures to gain a more complete understanding of computational energy consumption. In the future, researchers will identify other important computational workloads at NREL, such as computational fluid dynamics and wind resource forecasting, and build similar data sets to identify more efficient approaches.

This fundamental work will help establish energy efficiency optimization of different computational approaches, algorithms, and data management practices as a long-term capability in the laboratory.

“Investment in computing has traditionally focused on using it as a vehicle to conduct more complex clean energy research,” Tripp said. “But we have to start looking at computing as its own challenge and area of ​​energy efficiency research.”

For more information, read about the Green Computing Catalyst and NREL high performance data centers.

Leave a Reply

Your email address will not be published. Required fields are marked *