<aside> ✴️ Directory for Planetary-Scale Computation: An industry primer on the hyperscale CSP oligopoly (AWS/Azure/GCP):
Let’s Get Physical, (Cyber)Physical!: Flows of Atoms, Flows of Electrons
A Cloudy History: Four Histories of Cloud Computing
Primer on the Economics of Cloud Computing
Three-Body: Competitive Dynamics in the Hyperscale Oligopoly
[WIP] The Telos of Planetary-Scale Computation: Ongoing and Future Developments
Appendix:
[WIP] Clouds with Chinese Characteristics
[WIP] Deployment Models: Private/Hybrid/Multi-Cloud and Edge
[TBD] Green Clouds
[TBD] Netflix Case Study
[TBD] Snowflake Case Study
</aside>
<aside> ✴️ Table of Contents for Let’s Get Physical, (Cyber)Physical!: Flows of Atoms, Flows of Electrons:ss
</aside>
Not a cell phone in sight. Just computers living in the moment.
Not a cell phone in sight. Just computers living in the moment.
Needless to say, computers will not magically produce a quick technological fix to urban problems. For one thing, they may still evolve in the direction of routinization, further eroding the combinatorial richness of knowledge and making flows of information ever more sterile. The digital revolution should be thought of as one more element added to a complex mix, fully coexisting with older components (energetic and material), not all of which have been left in the past. In other words, digital machinery is simply a new node that has been grafted on the expanding autocatalytic loop. Far from having brought society to a new stage of its development, the information stage, computers have simply intensified the flow of knowledge, a flow which, like any other catalyst, still needs matter and energy flows to be effective.
— Manuel DeLanda, A Thousand Years of Nonlinear History (1997)
So, impulsively, I took a fiber-optic cable and unplugged it. Then I held it up to my eye and looked in. On the other side of the fiber, I imagined, was Japan. The light was red, and it winked like a star on a smoggy night. — Tung-Hui Hu, A Prehistory of the Cloud (2015)
<aside> ✴️ From An Insider's Look: Google's Data Centers (Cloud Next '19):
[2:30] **Joe Kava (**VP of Global Data Centers at Google): I remember the very first that I was at a Google data center and everyone kind of wonders what a data center is like at Google — I did too — when I got there, I was kind of tracing the path of the electrons in my mind ...
</aside>
I’ve never seen the Cloud — have you?
ASCENSION by @dominikus
<aside> ✴️ Resource List:
“The” [implied singular] “Cloud” is actually an assemblage of multiple “clouds” that are each materially composed of networks of hulking structures of earth, glass, and metal — elementally, the furthest possible things from actual clouds. If we define the “Cloud” as the aggregate mass of non-localized computers available globally, either in use or available for use by and from hyperscale providers, then, using data from Synergy Research Group, the “Cloud” was comprised of 659 large data centers as of Q2 ‘21 (more than doubling since mid-2016), 700 hyperscale data centers as of the end of Q3 ‘21, and 700+ “hyperscale” (more on this term later) data centers as of the time of your reading this.
The ever-increasing size of large-scale data centers range from common 300,000 sq. ft data centers to 10.7 million square feet (sq. ft) behemoths [for scale, Manhattan is 470.0 mm sq. ft & Central Park is ~39 mm sq. ft], as in China Telecom’s Inner Mongolian Information Park — each of these hyperscale data centers are home to tens of thousands of servers. If we assume ...
... then a rough estimate of the “Cloud’s” mass is 1,505,000,000 pounds or approximately 750,000 tons [for reference, the Empire State Building weighs ~365,000 tons], and that’s not even including the server cabinet, shelves, auxiliary equipment, batteries, etc. nor the actual building (or “shell”) that houses these server racks.