Editing Drum memory (section)

== Technical design ==
A drum memory or drum storage unit contained a large metal cylinder, coated on the outside surface with a [[ferromagnetic]] recording material. It could be considered the precursor to the [[hard disk drive]] (HDD), but in the form of a drum (cylinder) rather than a flat disk. In most designs, one or more rows of fixed [[read-write head]]s ran along the long axis of the drum, one for each track. The drum's controller simply selected the proper head and waited for the data to appear under it as the drum turned ([[rotational latency]]). Not all drum units were designed with each track having its own head. Some, such as the [[English Electric DEUCE]] drum and the UNIVAC [[FASTRAND]] had multiple heads moving a short distance on the drum in contrast to modern HDDs, which have one head per platter surface.

In November 1953 Hagen published a paper disclosing "air floating" of [[Disk read-and-write head|magnetic heads]] in an experimental sheet metal drum.<ref>{{Cite book |last=Hagen |first=Glenn E. |url=https://archive.org/details/sim_computers-and-people_1953-11_2_8/page/n25/ |title=Computers and Automation 1953-11: Vol 2 Iss 8 |date=1953-11-01 |publisher=Berkeley Enterprises |others=Internet Archive |pages=23, 25 |language=en}}</ref>  A US patent filed in January 1954 by Baumeister of IBM disclosed a "spring loaded and air supported shoe for poising a magnetic head above a rapidly rotating magnetic drum."<ref>{{cite web
 |url=https://patentimages.storage.googleapis.com/95/22/5b/71f30ff52b7f8a/US2862781.pdf
 |title=US Patent 2,862,781 RECORDING SUPPORT DEVICES  |last=Baumeister |first=H |date=December 2, 1958 |access-date=July 1, 2023}}</ref>  Flying heads became standard in drums and [[hard disk drive]]s.

Magnetic drum units used as primary memory were addressed by word. Drum units used as secondary storage were addressed by block. Several modes of block addressing were possible, depending on the device.
*Blocks took up an entire track and were addressed by track.
*Tracks were divided into fixed length sectors and addressing was by track and sectors.
*Blocks were variable length, and blocks were addressed by track and record number.
*Blocks were variable length with a key, and could be searched by key content.
Some devices were divided into logical cylinders, and addressing by track was actually logical cylinder and track.

The performance of a drum with one head per track is comparable to that of a disk with one head per track and is determined almost entirely by the rotational latency, whereas in an HDD with moving heads its performance includes a rotational latency delay plus the time to position the head over the desired track ([[seek time]]). In the era when drums were used as main working memory, programmers often did [[optimum programming]]—the programmer—or the assembler, e.g., Symbolic Optimal Assembly Program (SOAP)—positioned code on the drum in such a way as to reduce the amount of time needed for the next instruction to rotate into place under the head.<ref>
{{Cite manual |     title = SOAP II - Symbolic Optimal Assembly Program for the IBM 650 Data Processing System
 | publisher = IBM| id = 24-4000-0
 |       url = http://bitsavers.org/pdf/ibm/650/24-4000-0_SOAPII.pdf
 |      mode = cs2
 }}</ref> They did this by timing how long it would take after loading an instruction for the computer to be ready to read the next one, then placing that instruction on the drum so that it would arrive under a head just in time. This method of timing-compensation, called the "skip factor" or "[[interleaving (disk storage)|interleaving]]", was used for many years in storage memory controllers.