联邦能源研究：高密度、大规模计算与电网的交汇点.pdf

《联邦能源研究：高密度、大规模计算与电网的交汇点.pdf》由会员分享，可在线阅读，更多相关《联邦能源研究：高密度、大规模计算与电网的交汇点.pdf（35页珍藏版）》请在三个皮匠报告上搜索。

1、A Listening Session,LL21CFederal Energy R&D for Data CentersOctober 15,2025SpeakersMary Ann PietteAssociate Laboratory Director for Energy Technologies,Lawrence Berkeley National LaboratoryDale SartorLawrence Berkeley National Laboratory Affiliate and Founder,DOE/Berkeley Lab Center of Expertise for

2、 Data Center EnergyCosimo PecchioliIT&OEM Partnership Director,BP/CastrolArman ShehabiStaff Scientist,Lawrence Berkeley National LaboratorySarah SmithResearch Scientist,Lawrence Berkeley National Laboratory1.Identifying and prioritizing the most challenging technical barriers in supporting high-dens

3、ity compute.2.Identifying industry-wide practices and trends for calibrating industry-wide models of U.S.data center energy use.PurposeCoolingPoweringSelection of IT equipment by data center types and workloads DOE/LBNL data center overview Mary Ann PietteModerated discussion:Dale Sartor,Critical ad

4、vanced cooling Cosimo Pechiolliand power technologies Moderated discussion:Sarah Smith,Calibrating industry-wide models Arman Shehabiof U.S.data center energy useAgendaR&D:Funding R&D and testing of transformative cooling technologies Advancing microelectronics for efficient,high-density computingDe

5、veloping advanced computing and operating strategiesAdvancing large load flexibility and demand response for grid stabilityDOE Data Center PortfolioApplied Efforts:Opening federal lands and powerto advanced data centersAccelerating large grid projectsForging data center,utility,government partnershi

6、psProviding technical assistance to states,utilities,fed.data centers,including training on retrofits and cooling transitionsAnalysis:Market modelingGrid planning,including deployment and transmissionEmerging large load tariff structuresDOE Data Center PortfolioAccelerating AI at LBNLOptimization Te