Diana Marculescu: Computing for Sustainability

Diana Marculescu: Computing for Sustainability


My name is Diana Marculescu. I’m a professor of electrical and computer
engineering at Carnegie Mellon University. Most of my work relates to sustainability
in computing, but also more recently, computing for achieving sustainability in other fields. When something becomes more efficient, we’re
going to use more of it. Data from CISCO, for example, shows that for
each person, there’s at least two or three smart devices connected to the internet, and
that number is actually exceeding right now. So we’re not really going to achieve energy
efficiency per se if we’re actually going to use more of it, right? Because at this point, much of the energy
still comes from fossil fuels. So, one thing we’d like to do is use what
we have for something that has a larger social impact for aiding in renewable energy distribution
and generation. The concept of sustainability, meaning maintaining
forever a certain performance or behavior, actually exists in nature. So there is a certain jellyfish called Benjamin
Button jellyfish, because it actually can turn itself, under stress, from whatever shape
it is into multiple polyps that actually turn into young jellyfish, so in theory they could
live forever, unless they’re eaten by predators, so we’re trying to replicate that behavior
somehow. The same paradigm could use power-efficient
computing systems to aid in renewable energy generation, which in turn could help power
fabs that manufacture those exact systems. We all know about renewable energy generation,
so wind, solar, hydro. For wind power, we use arrays of turbines. For solar, we use arrays of solar panels,
but for water, we’re still in the hydro dam paradigm. So the idea is to move away from that and
try to use an array approach, where we could use these energy-efficient computing devices
to actually determine the best balance between the demand from the consumer and what we have
from the hydro power generation. So it’s the best marriage between achieving
sustainable computing and using those to achieve sustainability in renewable energy distribution
and generation. We just started this project. It’s funded by the National Science Foundation. We hope to have results in three to five years. We’re working with specialists from pricing
and energy and policy, as well as specialists from hydrology. So eventually, we’re hoping to show that this
is something feasible, so that could be used, for example, in addition to existing hydro
power projects, or new power projects in places where there wouldn’t be enough resources to
build a huge dam, and developing economies are one such example.

Leave a Reply

Your email address will not be published. Required fields are marked *