Examining how data centers work as sites of massive resource consumption integral to the cloud in ways that also restructure relationships between places and peoples.
Description:
Now that we have discussed the multiple ways in which the web is material, let’s examine resource consumption of data centers. The readings outline how data centers use water, energy, and other resources.
“The cloud has become so naturalized in everyday life that we tend to look right through it, seeing it uncritically, if we see it at all,” wrote Tung-Hui Hu notes in the Prehistory of the Cloud (2015: XII). How can we “see” the cloud? In Silvio Lorusso’s “Data Centers Grand Tour,” the viewer is shown images of data centers, and those data centers are the physical sites where the image is being “served” from. Lorusso seeks to underscore to viewers that data is stored somewhere. Data centers are a key element of what is sometimes labeled “the cloud.” As John Durham Peters notes in The Marvelous Clouds, “Media old and new are embedded in cycles of day and night, weather and climate, energy and culture, and they presuppose large populations of domesticated plants, animals, and humans, to say nothing of an old and cold universe” (2015: 377). In this unit, we seek to understand how data centers, as sites of media storage, are embedded in these cycles.
Data centers can be small and exist in the corners of office buildings or they can be massive buildings that occupy vast plains. Often described as covering multiple football fields, the newest innovations in data centers are described as “hyperscale.” Within data centers are computers that store and serve data. Data centers have a number of other impacts on the peoples near whom they are located, namely that they use great volumes of energy, space and water. As Greenpeace reports in “Clicking Green,” data centers can consume energy that is renewable and not renewable. Greenpeace makes it clear that while some companies such as Amazon consume “dirty” coal energy, other companies are working towards consuming “renewable” energy. In Pasek’s (2019) analysis of Microsoft’s work in this space, and in Microsoft’s report on their activities, it is clear that Microsoft has made progress towards their goals of consuming 100% renewable energy. Additionally, the most innovative data center operators have decreased the energy needed to store data, as Shehabi et al. (2018) show in their papers and reports. Yet, researchers have found that moves towards efficiency or renewable energy does not mean that the overall energy consumption by data centers has decreased.
While data centers allow for massive amounts of data to be stored remotely, people also might want to access this data immediately (think: “Netflix and chill”). Thus, data centers are often colocated with significant networking infrastructure or placed near consumers to ensure speedy delivery of bits. So Netflix’s Content Delivery Network relies on Amazon’s AWS cloud services, which are distributed around the globe, to ensure that there are “local” copies of movies that people want to watch. Resources such as datacentermap.com illuminate what data centers are where and what is either in the same building or nearby. Readings by Hogan (2015), Levanda and Mahmoudi (2019), and Johnson (2019) offer perspectives on data centers in Utah, the Pacific Northwest and Iceland, and how those data centers refigure relations between urban and rural, implicate water resources in national security, and marginalize places within imperial regimes.
Readings:
Hogan, Mél. “Data Flows and Water Woes: The Utah Data Center.” Big Data & Society 2, no. 2 (2015)
Anthony M. Levenda and Dillon Mahmoudi. “Silicon Forest and Server Farms: The (Urban) Nature of Digital Capitalism in the Pacific Northwest” Culture Machine (blog), April 2, 2019.
Johnson, Alix. “Data Centers as Infrastructural In-Betweens” American Ethnologist 46, no. 1 (2019): 75–88.
Edwards, Paul N. “Knowledge Infrastructures for the Anthropocene.” The Anthropocene Review 4, no. 1 (2017): 34–43.
Art:
Lorusso, Silvio. Data Centers Grand Tour: (this data belongs here) Exhibit at e-Permanent, 2013.
Low Power Magazine, “How to build a low-tech website?”
Optional Readings:
Velkova, J. “Data That Warms: Waste Heat, Infrastructural Convergence and the Computation Traffic Commodity.” Big Data & Society 3, no. 2 (2016).
Anne Pasek “Managing Carbon and Data Flows: Fungible Forms of Mediation in the Cloud.” Culture Machine (blog), April 2, 2019.
Reports:
Gary Cook, Jude Lee, Tamina Tsai, Ada Kong, John Deans, Brian Johnson, and Elizabeth Jardim. Clicking Green: Who is winning the race to build a green internet? Greenpeace.org, 2017.
International Energy Agency. Digitalization & Energy. Paris: OECD/International Energy Agency, 2017.
Arman Shehabi, Sarah Josephine Smith, Dale A Sartor, Richard E Brown, Magnus Herrlin, Jonathan G Koomey, Eric R Masanet, Nathaniel Horner, Inês Lima Azevedo, William Lintner United States Data Center Energy Usage Report. Berkeley, CA: Lawrence Berkeley National Laboratory, 2016.
Shehabi, Arman, Sarah J. Smith, Eric Masanet, and Jonathan Koomey. “Data Center Growth in the United States: Decoupling the Demand for Services from Electricity Use.” Environmental Research Letters 13, no. 12 (2018).
Industry resources:
Data Center Knowledge. [Trade press]
Microsoft data centers. The Carbon Benefits of Cloud Computing: A Study on the Microsoft Cloud. Redmond, WA: Microsoft. 2018.
Discussion Questions:
- From Levanda and Mahmoudi: “What inequities are arising in the uneven development of data infrastructures within and beyond cities? How might we extend analyses of data centers and data infrastructures to understand the relationship between computing and socio-natural change? And how might these [data center] mappings elucidate new areas for contestation and resistance? What are the possibilities for more sustainable and equitable alternatives in digital economies?
- Are Low Power Magazine’s low power websites an alternative to data centers? Why or why not?
- After interrogating the extractive logics which underpin data centers, Edwards (2107) reminds readers that the storage and use of data also allows people to comprehend climate change. The important use of computing resources by climate scientists potentially raises important questions: What are the “right” uses of computing resources? When is the use of computing resources justifiable from the perspective of resource consumption? When is it wasteful? Can traditional appraisal techniques from archival practice guide us forward? What could a decolonized approach to data management look like?
Exercise: Tracing bits
Almost all of us have some kind of presence on the web whether that is through a blog, a Github account, social media such as Insta, or a website or app that we have developed. The data from these entities resides in data centers. But as the above articles illustrate, data centers not only use energy but also generate heat, use water and refigure relations between urban and rural communities. The challenge in this exercise is to do an accounting of our online footprint (extra bonus if you want to work in searches and other online activity), focusing on where those bits, stored in data centers, are coming from. Use the mapping resources such as datacentermap.com and industry resources to try to discover where your data might be stored. You will want to consider both the geography of Once you have located some of the sites where your bits are coming from (geographically), also consider where the energy, water and heat are coming from and going to at these locations. Use the articles above describing data centers in the Pacific Northwest, Utah, and Iceland to try to understand how these data centers might be using resources. Who lives in these areas? What is it like to live with a data center nearby?
Exercise: Writing
Data centers have been depicted in movies such as 2001: Space Odyssey, The Matrix (thought this data center is powered by people), Oceans Eleven, Skyfall, and Tron Legacy. These data centers are sites of action and secrets. Write an essay about the internet as though you were inside a data center. What would it feel like to sit in a data center? What senses would be activated? What activity is visible and invisible? What would the building look like from the outside? How do you get into the data center? Who else is there?