Serendip is an independent site partnering with faculty at multiple colleges and universities around the world. Happy exploring!

End-user expectations

Anne Dalke's picture

Many of you probably saw (if you didn't, please read!) the lead article in this morning's NYTimes, Power, Pollution and the Internet, which makes it clear that our thinking we are being "green" in this class, by being paperless, is worse than an illusion; what we are actually doing, as we meet virtually each weekend, is helping to waste vast amounts of energy: "what’s driving that massive growth" is "the end-user expectation of anything, anytime, anywhere.”


Nan's picture

P.S.  This engineer would

P.S.  This engineer would love to hear your feedback.  It is an opportunity for us to speak into the larger community to scientists who are actually making choices regarding the direction of cloud technology that will affect us all.

Nan's picture

Thanks for pointing that out

Thanks for pointing that out to us.  I read and forwarded it to several people, including an engineer working on cloud systems.  I've got some charcoal.  Maybe we could find a cave wall.

Nan Valen's picture

(from an engineer I know who

(from an engineer I know who works with cloud systems)

"To close the article, Glanz writes: 'That’s what’s driving that
massive growth — the end-user expectation of anything, anytime,
anywhere' said David Cappuccio, a managing vice president and chief
of research at Gartner, the technology research firm.'We’re what’s
causing the problem.'

The implication I take from that: consumers' demands are driving
ever-growing and dirtier (ie diesel generators) energy demands. This
kind of suggests the notion that users (consumers at home or office
workers) are to blame but I don't really think that's true. What I
mean is that users are expecting reasonable things from the computing
resources - availability, functionality. I think the provider should
be responsible for figuring out how to provide the service in a
power-efficient way...but they will only do that if they have the
right incentives to do so.

So I think the dominant issue is the overall lack of sufficient
incentives for people and businesses to conserve power. There are
technological solutions to a lot of what is mentioned in this article.
Despite the skeptics, the cloud technology of "virtualization"
(mentioned in the article) is real and mature enough to play a part
now. Other technologies, like flash memory, will probably be able to
help even more in the relatively near future. But there's only a
modest amount of motivation for data center operators to worry about
this because ultimately power is pretty cheap...and "dirty" power is
still cheaper than "clean" power. I think this is quite an
indictment against our government - and by extension, our electorate.

This article highlights quite an unfortunate issue, because "cloud"
technologies have the potential to significantly reduce the power
required for the computing that people across the country and planet
want to do. This is because a cloud can be designed to run at much
higher utilization rates that people's computers at home could ever
reach. But as the article says, the industry isn't motivated (enough)
to fix the issues in how data centers have been run historically.

Anyway, I think my fundamental answer to these ideas is that we
have more a question of motivations and decisions than we have a
question of technology. We can produce wind and solar and nuclear
power today, sufficient to meet our needs, and we could create ways of
using energy more efficiently. But we don't have the collective
will...we refuse to use our government's capacity to create effective
change in this area, despite the enormous benefits it would have.

I'll admit, I'm pretty worried about how climate change is going to
impact the world. The debate about this in American politics is an
interesting (if horrifying) study in how people's interests change the
"reality" that they see. Conservatives have continued to deny climate
change science for long past any hint of a legitimate debate...but
why? Well, because there are interests involved -- letting go of old
industries (coal, oil), changing lifestyles (driving smaller cars,
riding bicycles or mass transit more, eating less beef), and accepting
a more interventionist government role (regulating carbon emissions,
maybe taxing them or cap/trade). Easier, apparently, to find some way
of denying the inconvenient truth. I find this so strange. I mean, I
want climate change to be false just as much as the next guy, but just
because I want it doesn't make it true. So how do so many people just
wish it away? There's an element of ignorance, and the plausible
deniability that comes with it: if I don't really understand the
science, maybe I can doubt that it is real & legitimate. Maybe one
skeptical scientist among thousands can be 'right', the lone voice of
truth. Anyway, it's scary, but also fascinating."