The application of higher-order PDA-based flow analyses to object-oriented languages enables comprehensive and precise characterization of program behavior, while retaining practicality with efficiency. We implement one such flow analysis which we've named OOCFA2. While over the years many advancements in flow analysis have been made, they have almost exclusively been with respect to functional languages, often modeled with the calculus. Object-oriented semantics--while also able to be modeled in a functional setting--provide certain structural guarantees and common idioms which we believe are valuable to reason over in a first-class manner. By tailoring modern, advanced flow analyses to object-oriented semantics, we believe it is possible to achieve greater precision and efficiency than could be had using a functional modeling. This, in turn, reflects upon the possible classes of higher-level analyses using the underlying flow analysis: the more powerful, efficient, and flexible the flow analysis, the more classes of higher-level analyses--e.g., security analyses--can be practically expressed. The growing trend is that smartphone and mobile-device (e.g., tablet) users are integrating these devices into their lives, in more frequent and more personal ways. Accordingly, the primary application and proof-of-concept for this work is the analysis of the Android operating system's permissions-based security system vis--vis potentially malicious applications. It is implemented atop OOCFA2. The use of a such a powerful higher-order flow analysis allows one to apply its knowledge to create a wide variety of powerful and practical security-analysis "front-ends"--not only the permissions-checking analysis in this work, but also, e.g., information-flow analyses. OOCFA2 is the first PDA-based higher-order flow analysis in an object-oriented setting. We empirically evaluate its accuracy and performance to prove its practical viability. We also evaluate the proof-of-concept security analysis' accuracy as directly related to OOCFA2; this shows promising results for the potential of building security-oriented "front-ends" atop OOCFA2.

Research on network traffic measurement is becoming popular because it could find the reason and valuable information regarding the occurrence of timestamp accuracy error. In this thesis the performance of UDP traffic between application level and link level has been investigated on three different Smartphone operating systems. One Java ME traffic generator and sink was developed for this thesis work. Distributed Passive Measurement Infrastructure with Measurement Point was used for the experiment. To get high timestamp accuracy DAG 3.5E card was used which was synchronized with GPS. Through the different tests under different protocol layers, it can be concluded that the Android mobile performs better as Java ME application works like a native application, whereas Symbian and Windows mobile struggles with incompatible implementation for poor JVM support. Time stamping at link level is more accurate and far different from application level. This thesis also investigated the performance evaluation of different JVM for Windows mobile.

Řada programovacích jazyků byla schopna zvýšit svoji rychlost výměnou běhových systémů stavěných na míru za obecné platformy, které pro optimalizaci používají just-in-time překlad, jako jsou GraalVM nebo RPython. V této práci vyhodnocuji, zda je použití takovýchto platforem vhodné i pro jazyky se závislymi typy nebo důkazovými systémy. Tato práce představuje koncepty -kalkulu a teorie typů potřebné pro úvod do závislých typů s relevantními algoritmy, specifikuje malý závisle typovaný jazyk založený na $\lambda\Pi$ kalkulu, a prezentuje dva interpretery tohoto jazyka. Tyto interpretery jsou psané v jazyce Kotlin, první je jednoduchý, psaný ve funkcionálním stylu a druhý používá platformu GraalVM a Truffle. GraalVM je platforma založená na virtuálním stroji Javy (JVM), která přidává just-in-time překladač založený na částečném vyhodnocení (partial evaluation) a Truffle je knihovna pro tvorbu programovacích jazyků využívající tento překladač. Závěr práce vyhodnocuje běhové charakteristiky těchto interpreterů na různých zátěžových testech.Závěry práce jsou ale silně negativní. Vliv JIT překladu není znatelný ani přes snahu optimalizovat běžné algoritmy z teorie typů, které jsou zjevně nevhodné pro platformu JVM. Práce končí návrhy několika navazujících projektů, které by lépe využily možnosti Truffle a které by byly vhodnější pro implementaci závisle typovaných jazyků.

This master's thesis describes the analysis, design and implementation of an application for automatic generation of UML class diagram. Application is designed as a web service, which provides remote access, especially permanent actuality of generated class diagram. Input of the service is a compiled application written for C# .NET or Java platform. The reader is acquainted with basics of reverse engineering of mentioned platforms and with structure of UML class diagram. Then are these knowledge applied in design and implementation of the service. The main goal is to facilitate and accelerate the activities of software development team members.

This project addresses parallelization of building and testing projects written i Java programming language. It proposes software that uses methods of continual integration, parallelization and distribution of computationally intensive tasks to grid architecture. Suggested software helps to accelerate the development of software product and automation of its parts.

The Ericsson AXE-based systems are programmed using an internally developed language called Plex-C. Plex-C is normally compiled to execute on an Ericsson internal processor architecture. A transition to standard processors is currently in progress. This makes it interesting to examine if Plex-C can be compiled to execute on the JVM, which would make it processor independent. The purpose of the thesis is to examine if parts of the run-time environment of Plex-C can be translated to Java and if this can be done so that sufficient performance is obtained. It includes how language constructions in Plex-C can be translated to Java. The thesis describes how a limited part of the Plex-C run-time environment is implemented in Java. Optimizations are an important part of the implementation. It is also described how the JVM system was tested with a benchmark test. The test results indicate that the implemented system is a few times faster than the Ericsson internal processor architecture. But this performance is still not sufficient for the JVM system to be an interesting replacement for the currently used processor architecture. It might still be useful as a processor independent test platform.