Carl Adam Petri
Carl Adam Petri ( born July 12, 1926 in Leipzig, † July 2, 2010 in Siegburg ) was a German mathematician and computer scientist. He is best known for the eponymous Petri nets for modeling distributed systems.
Life and work
Carl Adam Petri was born on July 12, 1926 in Leipzig. Petris father, a PhD in mathematics, led his son as a child in mathematics and inspired him for the natural sciences. So he got to the 12th birthday of the bankruptcy estate of a bookstore two thick textbooks of chemistry. At the age of 13 years, invented the Petri Petri nets with their graphics and rules for describing chemical processes. Through the relationship his father had Petri access to the central library in Leipzig with the (then banned ) works of Einstein and that of Heisenberg. In 1941, he learned from his father by Konrad Zuse and calculators whereupon he employed whose work with the physical laws for the purpose of computing and the machines and even built a small analog computer.
1944 Peter put down his Notabitur at the Thomas School in Leipzig and was pulled shortly thereafter to the military. He was Flakhelfer the Air Force and got into British captivity. Even during the captivity itself Petri dealt with the differences between the analog and the digital computer. He succeeded to the conclusion that digital computation methods are more versatile and more reliable. After the war he remained until 1949 in England and worked on solutions of surveying problems ( for example, the system of concentric ellipses on hilly terrain ). His reflections on the calculators but he did not give up.
1950 Peter returned to Germany and began to study mathematics at the Technical University of Hanover. There he received, among other things, a scholarship of the German National Academic Foundation. The just -developing computer science employed him on, because in 1955 he attended a training at IBM. After 1956 he received his diploma in mathematics, he became a research assistant at the Technical University of Hanover and later at the University of Bonn. 1962 Peter was awarded his doctorate of science at the Technical University of Darmstadt. The title of his dissertation is "Communication with Automata " and it deals with, among other things, simultaneous models ( Petri nets ).
From 1963 taught Peter the data center of Bonn University and forwarded this to 1968., Where he had the opportunity to continue working on his network theory, which was associated to some extent with the practice of the data center. After the founding of the Society for Mathematics and Data Processing ( GMD) 1968, he was the Research Institute for Information Systems, and brought this up to 1991. During this period resulted in collaborations with many European countries, the USA, India, Chile and China. After his retirement in 1991 he built his ideas further and published them.
Dissertation communication with machines
The title of the dissertation is designed deliberately ambiguous: On the one hand it can mean, as a man with a machine to communicate ( Turing's thought experiment: Turing test ) or the communication between people with the help of machines. Peter himself has stated that the second meaning of the title is intended to be his thesis but mainly dedicated to the first interpretation.
The work Petris is not a conventional dissertation, is in the solved an open problem or presented a new theory and worked out. As in many of his later works also, Petri formulated a wealth of suggestions and proposals for a new basis of theoretical computer science, which are more like sketches of a research program. His work begins with a very specific problem that occurs in the mechanical calculation of recursive functions. With these functions, it is generally not possible to predict how much space they need for the calculation. It can "try out" on whether the function is terminated and the computational resources are sufficient. If this is not the case, you must restart the calculation with more resources. Petri is at this point the question whether it is not possible, the computing system to add more resources, then simply continues to expect. If you can realize such a computer architecture that can be expanded? The conventional architecture comes to their limits. Therefore Petri proposes an entirely new architecture. It stipulates that the system consists of individual components and therefore each operate independently asynchronously. Many of the individual components then yield a large, expandable system with additional components. The draft asynchronously operating cellar memory he wants to show that asynchronous systems are more efficient than synchronous.
Petri further argued: It is inappropriate to build the computer science on sequential models. Instead associate events of the time, you should assign cause-and- effect relationship after. In addition, one should give up the idea of global states, such as those used in the sequential models. In the real world an action change only a few components, but not the entire system. For illustration and modeling asynchronous systems Petri graphs used in this work with special notations and rules that Petri nets.