Magnetic core memory
Introduction
Magnetic core memory, often referred to simply as core memory, was a dominant form of random-access memory (RAM) for computers from the mid-1950s until it was gradually replaced by semiconductor memory in the 1970s. This non-volatile memory system was pivotal in the development of early computers, providing a reliable and durable means of storing data. Core memory operates by storing data in the magnetic state of small ferrite rings, known as cores, which are threaded with wires to read and write information.
Historical Context
The development of magnetic core memory was a significant milestone in the history of computing. Prior to its invention, computers relied on less reliable and slower forms of memory, such as Williams tubes and delay line memory. The invention of core memory is credited to An Wang and Jay Forrester, who independently developed the technology in the late 1940s and early 1950s. Wang's patent for the "pulse transfer controlling device" and Forrester's work at MIT's Whirlwind project laid the groundwork for the widespread adoption of core memory.
Technical Description
Structure and Composition
Core memory consists of tiny toroidal ferrite cores, each about 1-3 mm in diameter. These cores are arranged in a grid pattern, with each core representing one bit of data. The cores are threaded with three wires: the X and Y wires for selecting the core, and the sense/inhibit wire for reading and writing data. The cores are made of a ferromagnetic material that can be magnetized in either of two directions, representing binary 0s and 1s.
Operation
Core memory operates on the principle of magnetic hysteresis. To write data, a current is passed through the X and Y wires, creating a magnetic field that sets the core's magnetization. The sense/inhibit wire is used to control the direction of magnetization, allowing for the writing of either a 0 or a 1. Reading data involves passing a current through the X and Y wires to determine the core's magnetic state. If the core's state changes, a voltage is induced in the sense wire, indicating the stored bit.
Non-Volatility and Durability
One of the key advantages of core memory is its non-volatility. Unlike modern DRAM, core memory retains its data without power, making it highly reliable for early computing systems. Additionally, core memory is resistant to radiation and electromagnetic interference, which was particularly valuable in military and aerospace applications.
Applications and Impact
Core memory was used extensively in early computers, including the IBM 1401, PDP-8, and the Apollo Guidance Computer. Its reliability and speed made it the memory of choice for mainframes and minicomputers during the 1960s and early 1970s. Core memory's robustness also made it suitable for use in harsh environments, such as space missions and military applications.
Transition to Semiconductor Memory
The transition from magnetic core memory to semiconductor memory began in the late 1960s with the advent of integrated circuits and MOSFET technology. Semiconductor memory offered several advantages over core memory, including faster access times, lower power consumption, and greater storage density. By the mid-1970s, semiconductor memory had largely supplanted core memory in new computer designs.
Legacy and Influence
Despite being obsolete today, magnetic core memory played a crucial role in the evolution of computing technology. It provided a reliable and efficient means of data storage during a formative period in computer history. The principles of core memory, particularly its grid-based architecture and non-volatile nature, continue to influence modern memory technologies.