Authors: Jukka-Pekka Partanen (CSC - IT Center for Science ltd.), Esa Heiskanen (CSC - IT Center for Science ltd.), Aaron Andersen (National Renewable Energy Laboratory (NREL))
Abstract: Efficient energy usage of data centers has attention locally, nationally and globally. Many data centers are increasingly interested in utilizing waste heat reuse. Two organizations; CSC and NREL will provide an overview of the cooling and heat reuse processes with lessons learned from design, construction and operations.
The session will outline the metrics (ERF, ERE, CoP etc.) used and foster discussion of standards, gaps and the different approaches. Both sites will highlight metrics, methodologies and how differences affect the calculations.
Audience discussion and Q&A is aimed at engaging the community to understand potentiality for new waste heat reuse projects.
Long Description: Session leader Veli-Antti Leinonen, Specialist (CSC)
CSC Presenter: Esa Heiskanen, LUMI Operations Manager
NREL presenter: Aaron Andersen, NREL Advanced Computing Operations Manager
Goals:
Our goal for hosting this BoF session is to engage the broader HPC community showcasing two different use-cases and setups for HPC heat reuse and how to operate, measure and monitor these environments.
Presenting two setups will help us to demonstrate the benefits, lessons learned and challenges of both designs, as well as, to highlight difference of included values when calculating the metrics. For example, the PUE value can increase if the heat pumps are owned and operated by the data center operator.
The HPC community is eager to understand the costs, complexity and benefits for waste heat utilisation. Lessons learned and understanding of impact to operational workload is also of interest.
Estimating the minimum size for the HPC load or other demographics suitable to evaluate the minimum size to start waste heat reuse and the most suitable cooling technologies.
Topic:
Examples of two heat reuse processes and how to measure the process with industry standards. Use of OCP heat reuse map to illustrate the locations that are already utilising heat reuse. (https://www.google.com/maps/d/viewer?hl=en&hl=en&mid=1bTp4Ugy7FGwfPadNlmfZpYwGY5Z5B7o&ll=28.871726229013795%2C-40.21607810348212&z=3)
Presentations will have following structure:
1. Introduction of the HPC site
2. Cooling process & heat reuse overview
3. Implementation of industry standards & metrics
4. Lessons learned
Relevance:
The audience will have clear understanding about the benefits and challenges on the showcased options, session will enable discussion around generalisation of waste heat utilisation and answer questions from the audience. At the moment there are a small number of HPC centres who utilise waste heat but several in planning stages.
More broadly cloud providers and commercial data center operations are looking for guidance to meet sustainability goals or in some cases legislation. Generalisation of liquid cooling (warm water cooled) compute in HPC enables wider scope of waste heat reuse than traditional air-cooled hardware, sharing operational experience is information that is hard to receive from commercial operators and thus, making the session a great value to SC 2023 guests.
Expected outcome:
Audience will have a clear view and scope features that affect waste heat cooling process in HPC setup. Presentation of process overview and clarification of two different environments. Presenters will share their experiences in building and operational phase and address the top-3 considerations that are the key for successful implementation of waste heat reuse in HPC environment.
Showcasing two real case-studies of how to implement industry standards in presented environments to measure the process and enable collection of data to fulfil the European Energy Efficiency directive of data center reporting.
Enable engaged discussion around the topic and opportunity to address questions directly to presenters.
Back to Birds of a Feather Archive Listing