Science Library - free educational site

Computer Hardware


Motherboard: the principle circuit board on which the essential computer components are centralised

The common components which make up personal computers are often provided on a single, exchangeable circuit board, known as the motherboard. Being on a single printed circuit board ensures better integrity of connections, as well as advantages in speed and compactness.


CPU, central processing unit, the primary component of a computer which controls the flow and operations on data

The Central Processing Unit (CPU) is the central set of components of a computer. The Von Neumann architecture of the CP comprises the control unit, for the control of the flow of data between the CPU and memory, and the arithmetic logic unit, for operations on data.

GPU: graphics processing unit

GPUs are used in a broad range of electronic devices to speed up the processing of large graphics files and their rendering.

GPU embedded on motherboard

Personal computers can have GPUs on separate video cards, or they can be integrated into motherboards. GPUs are also found as embedded systems in mobile phones, workstations and gaming consoles.

Although now a generic term, it was invented in a 1999 Nvidia launch of a single-chip processor with enhanced manipulation and rendering capabilities. Similar technology goes under the term VPU, the visual processing unit.

GPU on separate graphics card

GPUs can be dedicated graphics cards, with its own memory, which interface with the motherboard via an expansion slot. These are typically PCIe (Peripheral Component Interconnet express) and AGP (Accelerated Graphics Port).

GPUs may also be an integrated graphics processor (IGP), which use the computer's RAM. Nearly all commercially available computers come equipped IGP capability. These can cope with general graphics requirements, but heavy-duty games and graphically-demanding applications will require a dedicated card.

areas of application

3D Computer graphics

Matrix and vector non-graphical calculations


Video decoding

Motion compensation


Unlike RAM, Read-Only Memory (ROM) is designed not to be changed on a regular basis, if at all. Its most common application is firmware, software 'embedded' in devices essential for operation, and which does not require upgrading.

Integrated circuits (IC), diode matrix and mask ROM, have essential, unchangeable software. Recently, the definition of ROM has extended to include memory which is unchangeable in normal operation, but may still leave open possibilities for changes.

EPROM (erasable programmable read only memory), such as Flash ROM, as the name suggests, may be electrically erased. ROM may therefore refer to a non-volatile memory for the purpose of storing code and data.


Hard disk
RAM, read-access memory, a volatile memory used for rapid, temporary operations on data. Used in conjunction with more permanent storage devices, such as the hard disk.

RAM is a form of computer memory (data storage). Nowadays it usually takes the form of an integrated circuit and is most commonly found in desktop computers, laptops, gaming consoles etc.

    USB Flash drive interior circuitry
  • SRAM
  • Static Random-Access Memory

  • DRAM
  • Dynamic Random-Access Memory

  • Double Data Rate Synchronous Dynamic Random-Access Memory

  • ROM
  • Read Only Memory

  • HDD
  • HDD: Hard Disk Drive

  • Erasable Programmable Read Only Memory

  • Electrically Erasable Programmable Read-Only Memory


What defines and gives shape to a super computer facility is not the physical space for its arrays of ever more minituarized components, but that required for the infrastructure required to cool them. With enough heat to melt the expensive equipment in literally a flash, it needs the thermal sink capacity of a cold Alpine lake to keep it alive!

CSCS Swiss National Supercomputing Centre, Lugano

Computing Centre Lugano
The CSCS Super Computer Centre of Lugano started operations in its new facilities in March 2012

At Lugano's CSCS, a metre-diameter pipe draws water at 760 litres per second from thirty-five cool metres below the surface of Lugano's lake. This water, a numbing 6 degrees celcius in all seasons, is piped 2.8 km to the enormous, airline terminal-like building which houses the pumps and distribution network of pipes. The water is distributed to the various heat exchangers, which blow thusly refrigerated air past the components. The electron-heated water is then returned to the lake some 19 degrees the warmer for its experience.

The CSCS is Switzerland's new impressive centre for number crunchers extraordinaire. It is at the forefront of Swiss domestic needs, as well as many international projects, such as CERN and the IPCC's climate change modelling. All of the Swiss meteorological data is processed here, in an array of components that would fit into a smallish suitcase, and the universities of Switzerland, Germany and Britain vie for processing time. Applications for HPC (High-Performance Computing) range from the sciences to humanities, as well as commercial applications, such as modelling molecules for the pharmaceutical industry.

Lake water distribution
The high-performance computers need a constant flow of water at 6 degrees celcius to keep their components frizzle safe

Perhaps the most impressive thing about such a centre is its ability to channel, coordinate and reassemble the endless flow of data from huge projects, which require the power of a world network of computing centres. With many gigabytes per second of data needing to be processed and organised meaningfully, this demands a surrounding team of experts to design, implement and monitor all the activities. A cacophony of user demands occupy the fifty full-time staff, and all-in-all a third of a billion CPU hours are managed annually.

Swiss Meteological Service
The Swiss Meteorological service crunches all of the data from its satellite and terrestrial sensors in this tiny cabinet

The first supercomputer used in any Swiss supercomputing centre was installed in 1992, and had a capacity of 0.005 teraflops, or a wimpy five billion floating point operations (techy speak for calculations) per second - a task your PC could manage with or without its morning coffee. It took ten years before a supercomputer broke the threshold of one terabyte of floating point operations every second. Another decade along, and CSCS's Monte Rosa CRAY XE6 can manage over 400 of these teraflops. Eloquent proof of the veracity of Moore's Law.

The most recent addition to the CSCS is the Piz Daint, following the budding tradition here named after a mountain in Switzerland. Piz Daint has 1,256 compute nodes, each with two 12-core Intel® Haswell CPUs (Intel® Xeon® ES-2690 v3), providing 1.192 nodes with 64 GB of RAM each, and 64 compute nodes with 128 GB RAM each. The array has 7 cabinets of Cray XC40, totalling 76 TB of memory, and a 1 PB scratch file system.

Piz Daint
The Piz Daint (Cray XC30) is the most powerful supercomputer in Europe, and can process at 7,787 teraflops (7.787 million billion floating point operations per second)

This is the most powerful supercomputer in Europe.

We might be tempted to marvel at this. At half a million billion operations per second per machine, we may prosaically come to the conclusion that this brings potential modelling resolution to grains of sand in the universe level. But just to give a perspective - and add a dash of awe for nature - despite this overwhelming capacity to calculate, if each operation were to represent a single atom of hydrogen, it would still take an entire year of constant operation to process but a single tonne of star stuff!

And of course, computing power will only ever be as good as the science behind it.....

Andrew Bone

Monte rosa
The Monte Rosa CRAY XE6 can process at 400 teraflops (400,000 billion floating point operations per second)

History of Computing Devices

Sumerian Clay Tablet

Dating back to c.2600 BCE, a mathematical table was found the Sumerian city of Shuruppag. It provided a list of length measures and their sum, supplying the user a ready look-up table of areas.

Salamis Tablet

A counting board from c. 300 BCE, from the Greek island Salamis.


similar in function to the Salamis Tablet, the abacus (Greek 'abax' or 'abakon' = table or tablet), is a counting board for the four basic arithmetic calculations (+, -, x, ÷). Beads, representing unit values, are moved on rods representing decimal places. The earliest known forms of abacus date back to 300 BCE in Sumeria, and 190 CE in China.

Opus Palatinum de Triangulis

Palatine Work on Triangles, 1596 book by Rheticus (1514-1574), the publisher of Copernicus' book On the Revolutions of the Heavenly Spheres, provides lists of trigonometric tables.

Napier's Rods

1617, Napier's Rods (or Bones) are a mechanical aid to multiplication.

Napier's Logarithms

John Napier (1550-1617) laboriously created a set of tables, approaching 10 million entries, for astronomers. the formulation was refined by Henry Briggs (1561-1639).

Slide Rule

1620: Edmund Gunter (1581-1626) created a logarithmic scale for calculations using a set of dividers.

William Oughtred (1574 - 1660) adapted this idea and used two logarithmic scales sliding against each other to perform multiplication and division operations. This was the original slide rule. He also invented a circular slide rule.

Pascal's Calculator

Blaise Pascal invented this mechanical calculator in 1648, after much trial and error.

Jacquard's Loom

Joseph Marie Jacquard invented a mechanical loom in 1801 which used a sequence of punched cards to programme a loom with a pattern. Ada Lovelace saw the potential of this principle for the development of software for use with Babbage's 'Analytical Engine'.

Stepped Reckoner

Leibniz adapted Pascal's calculator, and developed a machine in 1671-1673 which utlised s stepped drum. It was designed to perform all four arithmetic operations, but had difficulties in practice with its gearing mechanism.

Difference Engine

Charles Babbage developed a machine which, using a handcrank, could turn columns of numberwheels, to make calculations, and print out the results. He proposed a more advanced model, the Analytical Engine, which would have been a programmable, logic-gate machine, but failed to gain the funding necessary. Ada Lovelace developed software systems for this machine, and is considered the first software engineer.


Charles Xavier Thomas de Colmar developed the arithmometer, 1851, which used an accumulator to extend addition and subtraction to multiplication and division. It was used commercially till 1915.

Curta Calculator

Curt Herzstark released the Curta in 1948. It was a handcranked digital mechanical calculator. It could do the four arithmetic operations, and others, such as square roots.

Pocket Calculator

The first electronic pocket calculator (LE-120A Handy) was released in 1971 by the Japanese firm Busicom. Hewlett-Packard released the HP-35 scientific calculator soon after.

Content © All rights reserved. Created : November 9, 2014 Last updated :March 7, 2016

Latest Item on Science Library:

The most recent article is:


View this item in the topic:

Vectors and Trigonometry

and many more articles in the subject:

Subject of the Week


'Universe' on covers astronomy, cosmology, and space exploration. Learn Science with


Great Scientists

Charles-Augustin Coulomb

1736 - 1806

Charles-Augustin de Coulomb, 1736 - 1806, French physicist, whose name is the unit for electrical charge.

Charles-Augustin de Coulomb, 1736 - 1806, French physicist

Quote of the day...

Read, read, read, read, read, read... if you do not read, you will never make a great film.

ZumGuy Internet Promotions

Yoga in Mendrisio