Hungarian pages

Correctness of Distributed Functional Programs -- OTKA T037742

Contents

Back to the top of this page

Abstract

Nowadays distributed systems form one of the most important research areas in informatics. The appearance of the Java programming language, the CORBA standard and other software tools made it possible to utilize the object-oriented software technology also in the development of distributed applications. Hence the design and implementation of distributed systems came into practice. The components of such applications run concurrently, and can collaborate with one another by not only synchronization and communication of data, but also by communication of mobile code. The correctness of programs written in a functional language can be proven more easily than that of programs written in the more widely used imperative languages. In a pure functional language, referential transparency allows many program properties to be proven by mathematical induction. The behaviour and the state changes of intricate distributed systems can be described without the violation of referential transparency in modern functional languages like Haskell, Clean, etc. Thus the proof and the verification of the safety properties and the invariants of such systems can be completed more easily.

Our goal is to extend mobile code (which is already equipped with certain type information) with further semantical information concerning its behaviour, and to make the run-time verification of those properties possible. We are going to let a proof system examine the code of a Clean program before sending it to another host, and to bundle the properties---and also the main steps of the proof of the properties---with the code. Thus the receiving party can easily and efficiently check whether the code indeed possesses the supposed properties, so it can be sure that the safety properties of some program components or of the whole distributed system will not be violated. In this research we have to define the appropriate programming language constructs that allow of the description of different safety properties, we have to develop the methodology of proving these properties, automating the proving process as much as possible, and checking the proofs during run-time.

Back to the top of this page

People

Supervisor Dr. Horváth, Zoltán hz@inf.elte.hu http://people.inf.elte.hu/hz
Participants Dr. Csörnyei, Zoltán csz@inf.elte.hu http://people.inf.elte.hu/csz
Daxkobler, Károly mzperx@inf.elte.hu
Diviánszky, Péter divianp@delfin.klte.hu
Hegedűs, Hajnalka heha@inf.elte.hu
Hernyák, Zoltán aroan@inf.elte.hu
Dr. Kozma, László kozma@inf.elte.hu
Kozsik, Tamás kto@inf.elte.hu http://people.inf.elte.hu/kto
Dr. habil Pásztorné Varga, Katalin pkata@inf.elte.hu http://people.inf.elte.hu/pkata
Szabóné Nacsa, Rozália nacsa@inf.elte.hu
Tejfel, Máté matej@inf.elte.hu http://people.inf.elte.hu/matej
Zsók, Viktória zsv@inf.elte.hu

Back to the top of this page

Preliminary Results

Back to the top of this page

Results

Papers

2002

Back to the top of this page

2003

Back to the top of this page

Software

2003

Back to the top of this page

2007

Back to the top of this page

Reports, Textbook Chapters, Notes

2002

Back to the top of this page

Thesis for National Scientific Students' Associations Conference

2002

Back to the top of this page

Thesis

2002

Back to the top of this page

2003

Back to the top of this page

Lectures, Example Programs

2002

Back to the top of this page

2003

Back to the top of this page

Related projects

Back to the top of this page


Valid CSS! Valid HTML 4.01!

Magyar nyelvű oldalak

fun_ver

Last modified: Fri Feb 29 15:03:01 2008