#### Carver Mead (CalTech Professor) – Insight 1 (May 2018)

#### Chapters

#### Abstract

Fluid Dynamics: An Innovative Lens for Understanding Information Processing

In the rapidly evolving world of technology, understanding the intricacies of information processing is vital. This article aims to shed light on the fascinating analogy between fluid dynamics and information processing, offering a fresh perspective on how digital information is stored, processed, and distributed. By comparing the flow of water in a system of tanks and valves to the flow of electrons in silicon-based systems, we uncover the fundamental principles of computation and the critical aspects of energy and time costs involved in information processing.

The Essence of Information Processing: A Physical Perspective

At the core of information processing lies the necessity to give information a physical form. This can be in various mediums – from radio signals and magnetic regions on disks to more tangible forms like scratches on CDs or paper. In the field of electronics, the behavior of electrons in a conductor is particularly noteworthy. They exhibit fluid-like properties, which opens up an intriguing avenue for representing information in a fluidic manner.

Digital Information and the Water Tank Analogy

Bits, the fundamental units of information, can take various physical forms, such as radio signals, scratches on paper, magnetic regions on disks, or scratches on CDs.

A compelling way to visualize digital bits is through the water tank analogy. Imagine a tank with two valves – one at the top and one at the bottom. When the top valve is open, allowing the tank to fill, it signifies a logical one. Conversely, opening the bottom valve to drain the tank represents a logical zero. This simple yet powerful analogy mirrors the binary nature of digital information.

The Costs of Transitioning States

Every shift from one to zero (or vice versa) in this fluid-based system incurs energy and time costs. These costs are akin to the water moving through the system – from the trough to the tank, and then to the gutter. The size of the tank and the flow rate through the valves significantly influence the time required for these transitions.

The Challenge of Information Distribution

Distributing information in this model necessitates running a pipe to the intended destination, which introduces additional time and energy overheads due to the need to fill both the pipe and the tank. This aspect is crucial in understanding the challenges faced in information transmission, be it in fluidic systems or electronic ones.

Silicon-Based Information Processing: A Parallel

The principles of energy and time costs extend seamlessly to silicon-based information processing. In such systems, the presence or absence of charge in a silicon region represents a bit. Changing this charge state requires energy and time, paralleling the filling and emptying of the water tank in our analogy.

Memory and Logic in the Fluid Model

Digital information storage can be visualized as water trapped inside tanks by manipulating valves, a concept utilized in dynamic random access memories (DRAMs) in electronic devices. Furthermore, logic functions are executed using valves to control water flow. For instance, a tank with valves on either side can represent logical operations: if either valve is open, water flows in, denoting a logical “one”; if both are closed, no water flows, indicating a logical “zero.” By combining these bit representations through logic functions, universal computation is achievable.

Advanced Control: Valves and Water Pressure

To perform complex computations, the valve settings must depend on the water in other tanks. This is where water pressure comes into play, analogous to voltage in electronic devices. An electronic component that regulates the flow of electrons based on an external voltage is essential for replicating all the operations described in the fluid model using electronic means.

Bridging Fluid Dynamics and Electronic Computation

The water tank analogy serves as an intuitive and simplified model for understanding the core concepts of information processing. It highlights the significant energy and time considerations involved in manipulating and transmitting information. These insights are invaluable for designing and optimizing electronic systems capable of handling complex information processing tasks. By drawing parallels between fluid dynamics and electronic computation, we gain a deeper appreciation of the underlying principles governing the digital world.

Notes by: Random Access