Profile picture
Jonathan Koomey @jgkoomey
, 13 tweets, 4 min read Read on Twitter
Here we go again. So much wrongness in this article, it's hard to know where to start.
1) Anyone who extrapolates IT electricity use to 2040 (or even 2025) is making a grave error. Things change so fast in this field that you just can't make predictions that far out.
2) Growth in data use/transfers (even fast growth) does NOT necessarily imply growth in electricity use. Our 2016 analysis of US data centers showed almost a doubling of computations from 2010 to 2014 while electricity use stayed roughly flat. eta.lbl.gov/publications/u…
3) The data center industry is in a transition phase, moving away from traditional "in-house" corporate data centers to more reliance on much more efficient cloud based solutions, either "in-house" or in the public cloud.
4) There are good reasons why cloud is MUCH more efficient, as I explain here: gigaom.com/2011/07/25/4-r…
5) In house facilities can be improved, but the biggest effect will be a massive shift to cloud over the next 5 years. This may drive total data center electricity use DOWN, even with growth in computations and data transfers.
6) Internet of Things devices are often battery powered, so they need to be orders of magnitude more efficient than plugged in devices. They also help us measure and control much larger energy flows, so their net effect may be to reduce electricity use overall.
6) continued: Measuring those net effects is hard, but it's at least plausible to argue that the net effect of IT deployment will be reduced electricity use.
7) Shifting to cloud can be a cost saving option, but usually it requires some "pre-optimization", including removing zombie servers and consolidating low utilization instances. koomey.com/post/168267640…
8) There are many ways to conduct that optimization in corporate facilities, and all of those facilities should do it anyway, even if the shift to cloud isn't imminent.
9) Usually the impediments to the shift to higher efficiency are not technical but institutional and behavioral. Having separate budgets and separate bosses for different parts of data center operations creates principal-agent problems galore.
10) Continued: Also see datacenterdynamics.com/focus/archive/… for data centers and arjournals.annualreviews.org/eprint/wjniAGG… for general discussion of effects of IT on resource use.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Jonathan Koomey
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member and get exclusive features!

Premium member ($3.00/month or $30.00/year)

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!