[Congressional Record Volume 170, Number 126 (Thursday, August 1, 2024)]
[Senate]
[Page S5796]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]
SA 3222. Mr. HEINRICH submitted an amendment intended to be proposed
by him to the bill S. 4638, to authorize appropriations for fiscal year
2025 for military activities of the Department of Defense, for military
construction, and for defense activities of the Department of Energy,
to prescribe military personnel strengths for such fiscal year, and for
other purposes; which was ordered to lie on the table; as follows:
At the end of subtitle H of title X, add the following:
SEC. 10__. RESEARCH AND DEVELOPMENT ON HIGH-PERFORMANCE
COMPUTING AND ARTIFICIAL INTELLIGENCE ENERGY
EFFICIENCY.
(a) In General.--The Secretary of Energy, acting jointly
through the Under Secretary for Science and Innovation and
the Under Secretary for Nuclear Security, shall implement a
research and development program (referred to in this section
as the ``program'') to substantially improve the
computational energy efficiency of high-performance computing
and artificial intelligence at the Department of Energy and
the National Nuclear Security Administration.
(b) Target.--The program shall set a target of improving
the average energy efficiency of high-performance computing
and artificial intelligence computations by a factor of not
less than 10 during the period beginning on January 1, 2025,
and ending on December 31, 2029.
(c) Considerations.--The program shall take into account
all aspects of data center energy efficiency, including--
(1) software architecture;
(2) hardware architecture, including computation, memory,
and networking;
(3) data center design;
(4) electrical power generation, storage, and transmission;
and
(5) workflow management.
(d) Collaboration.--The Secretary of Energy shall--
(1) collaborate with industry partners from all aspects of
the high-performance computing and artificial intelligence
ecosystem in implementing the program; and
(2) to the maximum extent feasible, ensure that any
learnings from the program are shared with commercial vendors
in the high-performance computing and artificial intelligence
ecosystem with the goal of improving overall energy
efficiency of high-performance computing and artificial
intelligence computations in the United States.
______