Category: Data Center Energy
-
Overestimating data center public health costs
Energy & carbon aren’t the only environmental impacts of computing, but are the public health impacts overstated?
-
If only data centers would participate in demand response
Even for AI training workloads, data center demand response remains an academic exercise – intriguing but impractical.
-
Expect more overestimates of AI energy consumption
We’ve started to see AI doomerism spread to predictions of the vast quantity of energy AI is undoubtably going to consume.
-
Stretched grid? Managing data center energy demand and grid capacity
As more data centers are built, competing with other users like housing projects and electric vehicle (EV) charging, how should the energy system change to accommodate these evolving uses?
-
Paper Notes – The world wide web of carbon
Will we ever be able to accurately assess the carbon footprint of IT? Could a relational footprint methodology be more useful? Paper notes on Pasek et al (2023).
-
Improving the accuracy of the GHG protocol in time and space
The GHG Protocol needs to evolve to better represent the complexity of how energy markets work.
-
Influencing the carbon emissions of AI
There is a correlation between the training time and energy consumption, but that doesn’t mean there is a correlation between training time and carbon emissions.
-
Why don’t data centers participate in demand response?
Demand response is where a user of energy (usually electricity) varies its demand for a period of time in response to a request from the grid operator. This typically occurs when demand might outweigh supply. If it is not possible to increase supply in response to higher prices, the next…
-
Is carbon the right metric for developers to optimize?
Developers need metrics that are consistent and easy to optimize, but carbon intensity varies in time and space. So when is it a useful metric for developers to consider?
-
Paper notes – Mitigating Curtailment and Carbon Emissions through Load Migration between Data Centers
This is a good paper that makes valid points about the possibilities of migrating flexible IT workloads, however it makes classic assumptions I see in most papers that discuss this topic.