WEMO24-Compiled Report-Interactive compressed (002) 4154-7882-8116 v.1 - Flipbook - Page 107
The decentralization of compute power is poised to significantly
transform the data center model and its associated energy
dynamics.
One of the most prominent forms of decentralization is the
development of edge data centers. These smaller facilities,
located close to the network edge and, by extension, end-users
offers several important advantages, including the ability to be
powered by local renewable energy sources. This makes them
ideal for supporting advanced digital use cases, such as 5G
communications, which require high compute and low latency.
In addition, smaller, edge-based data sites can be more easily
integrated into local energy management schemes, allowing
for the reuse of waste heat, as well as better alignment
with renewable energy source availability and avoid grid
developments. This localized approach not only enhances
energy efficiency but also contributes to a more sustainable
and resilient energy ecosystem.
Data centers as an energy producer
The decentralization of compute power could lead to a parallel
decentralization of energy production. With respect to data
centers, this means developing their own electrical production
and/or storage capabilities.
With the potential to establish micro-electric power sites or
connect directly to renewable energy sources, data centers
could become more self-sufficient and reduce their reliance
on traditional power grids.
For example, in Ireland, new policies are being introduced that
would allow private companies, most notably, data center
operators, to build their own energy infrastructure. This could
provide a workaround for a de facto moratorium on data
center developments in the greater Dublin area that is in effect
through 2028.
Taking a holistic approach to energy
management
While the growing use of advanced digital technologies are
increasing energy consumption, these very same applications
can also have the potential to significantly enhance sustainability
efforts for energy companies.
For example, AI and generative AI can support the reduction of
electricity consumption through optimized energy management
and predictive maintenance. AI can also be used to analyze
vast amounts of data to identify patterns and inefficiencies
within energy grids, buildings, factories, and transportation
systems. This analysis can lead to smarter energy distribution
and consumption strategies, minimizing waste and lowering
overall electricity usage.
NEXT
ARTICLE
The “push and pull” nature of these technologies highlight
the need for organizations to take a holistic approach to
sustainability measurement, assessing both the direct and
indirect impacts of AI implementation. It also requires companies
to have a more complete understanding of the total impact of
new technologies.
HOME
Decentralizing compute power
PREVIOUS
ARTICLE
For example, optimizing the efficiency of EVs through AI
could lead to more extensive use of these vehicles, potentially
offsetting some of the initial gains in efficiency—a phenomenon
known as the rebound effect.
To truly enhance sustainability, energy companies must consider
the broader implications of AI and generative AI applications,
ensuring that the overall environmental impact remains positive.
Integrating AI thoughtfully into operations can help energy
companies achieve a more sustainable future while navigating
the complexities of energy consumption and ecological impact.
WEMO 2024
How to meet Data center AI/
Gen AI consumption increase?
107
08
