[Paper Review] Compute at Scale: A Broad Investigation into the Data Center Industry

Konstantin Pilz, Lennart Heim|arXiv (Cornell University)|Nov 5, 2023

Cloud Computing and Resource Management4 citations

TL;DR

This paper provides a comprehensive analysis of the global data center industry, examining its infrastructure, economic scale, and role in enabling large-scale AI workloads. It identifies key drivers such as power consumption, cooling, connectivity, and security, and estimates the industry's value at $250B with projected growth to $500B in seven years, supported by ~500 large data centers across the US, Europe, and China.

ABSTRACT

This report characterizes the data center industry and its importance for AI development. Data centers are industrial facilities that efficiently provide compute at scale and thus constitute the engine rooms of today's digital economy. As large-scale AI training and inference become increasingly computationally expensive, they are dominantly executed from this designated infrastructure. Key features of data centers include large-scale compute clusters that require extensive cooling and consume large amounts of power, the need for fast connectivity both within the data center and to the internet, and an emphasis on security and reliability. The global industry is valued at approximately $250B and is expected to double over the next seven years. There are likely about 500 large (above 10 MW) data centers globally, with the US, Europe, and China constituting the most important markets. The report further covers important actors, business models, main inputs, and typical locations of data centers.

Motivation & Objective

• To characterize the structure, scale, and economic significance of the global data center industry.
• To analyze the technical and operational requirements of large-scale data centers, including power, cooling, and connectivity.
• To identify major market regions, key players, and business models shaping the industry.
• To assess the industry's role as the foundational infrastructure for large-scale AI training and inference.
• To project future growth and investment trends in data center capacity and deployment.

Proposed method

• Conducting a broad empirical investigation into data center infrastructure using publicly available industry data and reports.
• Analyzing power consumption, cooling systems, and network connectivity requirements across large-scale data centers.
• Mapping the geographic distribution of large data centers (above 10 MW) across major global markets.
• Evaluating business models and key actors, including hyperscalers and regional providers.
• Estimating market size and growth using financial and operational benchmarks from industry sources.
• Synthesizing findings into a structured overview of data center operations and their role in AI infrastructure.

Experimental results

Research questions

• RQ1 What is the current global scale and economic value of the data center industry?
• RQ2 What are the primary technical and operational requirements for large-scale data centers, including power, cooling, and connectivity?
• RQ3 Which regions and countries host the majority of large data centers, and what drives their market dominance?
• RQ4 How do business models and key actors shape the evolution of the data center industry?
• RQ5 What is the projected growth trajectory of the data center industry over the next seven years?

Key findings

• The global data center industry is valued at approximately $250 billion and is projected to double in value within seven years.
• There are likely around 500 large data centers worldwide, each with a capacity exceeding 10 megawatts.
• The United States, Europe, and China represent the most significant markets for large-scale data center deployment.
• Data centers are essential infrastructure for large-scale AI training and inference, driven by high computational demands.
• Key operational challenges include managing massive power consumption, implementing effective cooling systems, and ensuring high-speed connectivity.
• Security and reliability are central design priorities in data center architecture and operations.