The elements of a physical IT architecture

Every IT system is designed for installation on a physical IT architecture. In order to describe physical architectures for IT systems this section uses a simple classification that exists of six elements: permanent storage (mostly disk space), temporary storage (also known as internal memory), the central processing unit (or: CPU), data entry devices, data output devices, and the data transportation network. Unlike the logical architecture of an ERP system, the physical architecture is not only conceptual. Permanent and temporary storage, CPU’s, data entry and output devices, and networks are tangible.

Two elements in a physical architecture take care of data storage. Firstly, there is permanent data storage, which secures that data are kept without the need for electricity or other power sources. In general, data are stored on disks, but tapes, CD’s, DVD’s and USB devices are also suitable for permanent data entry.

The enormous growth in permanent storage capacity in the past decades, combined with a fast price decrease has been remarkable. As an illustration: a hard disk with a capacity of 15 Megabyte cost €750 in 1986, while a USB stick with a capacity of 16 Gigabyte cost €8 in 2014. This means a price decrease from €50 000 to €0.50 per Gigabyte.

In addition to this substantial price decrease, the speed of storage on and retrieval from modern media has also increased enormously. However, for the fast storage and retrieval of data, permanent storage is not sufficient. Here, the second element of a physical IT architecture is applied: the temporary data storage.

Temporary data storage, also called internal memory, consists of memory cards. An enormous technological development has also taken place here: the first personal computers that were available on the market in the 1980s had an internal memory of 640 KiloByte, while a modern personal computer has at least a thousand times this amount. When compared to permanent storage, internal memory is fast, but also expensive. It is also more vulnerable for intentional and unintentional power interruption, as the data are lost when the computer is powered down.

The transaction processor in an IT architecture is normally indicated with terms like processor, central processing unit or CPU. A CPU processes information by applying calculations to it or by rearranging it. CPU’s consist of semiconductors on silicon wafers. The processing power of CPU’s has increased exponentially in the past decades. In 1965, Moore formulated his well-known law, which says that every year the processing power of CPU’s doubles [Moore, 1965]. This law still holds today, despite the fact that specialists have repeatedly announced that the end of the development is within sight. World-wide, the market leadership in the development and supply of  processors of the company Intel is undisputed.

The input and output devices are those elements in an IT architecture that take care of the exchange of information between people and computers. In the early years of computerization this exchange was realized by means of punched cards. On the data entry side, this inefficient way of data exchange was soon replaced by keyboards. Since the introduction of Windows the mouse has become an indispensible device for data entry. In the past years, a variety of data entry devices has been developed, such as bar code readers, text scanners, iris scanners, and electronic pens. In the future, these devices may all become obsolete, when speech recognition has become more sophisticated and data will be entered directly via the human voice.

On the data output side, that is information exchange from computers to people, punched cards were in the irst instance replaced by a computer monitor on which green characters were displayed on a black background. Today, computer monitors can display an infinite number of colors in very high resolutions. Other output devices that are available are printers, plotters, blackberries, smart phones and smart watches.

With the current technological level of devices, it is already possible to realise advanced data integration with customers, suppliers and other parties. Because of opportunities offered by technological innovation, data entry and output will be realised more and more without human intervention, which means that data entry and output devices will be used less in the future.

When the various elements of a physical IT architecture are on different locations, information has to be transported between those locations. Data networks take care of this. The capacity and speed of such networks have gone through enormous growth. It is not so long ago that a speed of 9.6 Kilobit per second over a telephone line was acceptable, while today a speed of 11 Megabit is not unusual. In parallel, the cost of network capacity has come down dramatically. A private company network with an availability of considerably less than 100 percent used to cost millions of Euros annually, while transportation of data over the Internet is now already possible for less than one hundred Euro per year. In below picture the Internet access per country in continental European countries in 2012 is presented [Seybert, 2012]. In some countries, more than 90 percent of individuals already use the Internet. If Internet access continues to grow, European citizens will be always on line in the near future.

 

As a result of enormous technological innovations, physical IT architecture with high capacity, quality and speed is available today at low prices. However, the computer applications that are installed on this architecture have also become more demanding. This certainly holds for ERP systems. It is therefore still relevant for every ERP implementation to make a well-designed mapping of the logical elements of the ERP system on the elements of the physical IT architecture.

In the next three sections examples are presented of three well-known IT architectures and the way ERP can be represented on them. It has to be noted that the physical architectures are described in their purest forms. In reality, these pure forms do not occur and mixtures of architecture types are being used.