- english version ZlomekFS is a distributed filesystem which is able to share data among nodes. These nodes can cooperate in various modes. Modes differ in caching of local data at the client side and in the way how data is synchronized between the client and the server. This synchronization has been implemented by an unsecured and untrusted network connection until now. The current implementation uses FUSE interface for the communication between the kernel of the filesystem and the user. The synchronization is triggered when the user performs a specific operation above the filesystem. For well functionality in modern network environment it is necessary to both communicating partners to be sure about the identity of the second end of a communicating channel. This thesis implements trustworthy connection between the server and the client part of filesystem. In addition, it separates machine and user type of client. Next important part of the thesis is correct data synchronization and precise definition of sharing semantics.

The use and application of embedded systems in everyday life has proliferated in the past few years. These systems are constrained in terms of power consumption, available memory and processing requirements. Typical embedded systems like handheld devices, cell phones, single board computer based systems are diskless and use flash for secondary storage. The choice of filesystem for these diskless systems can greatly impact the performance and the energy consumption of the system as well as lifetime of flash. In this thesis work, the energy consumption of flash based filesystems has been characterized. Both the processor and flash energy consumption are characterized as a function of filesystem specific operations. The work is aimed at helping a system designer compare and contrast different filesystems based on energy consumption as a metric. The macromodel can be used to characterize and estimate the energy consumption of applications due to filesystem running on flash. The study is done on a StrongARM based processor running Linux. Two of the popular filesystems JFFS2 and ext3 are profiled.

Many Online Web Services (OWSs) today, such as Flickr and Twitter, provide users with the possibility to post images that are stored on the platform for free. This thesis explores creating a cryptographically secure filesystem that stores its data on an online web service by encoding the encrypted data as images. Images have been selected as the target as more data can usually be stored in image posts than text posts on OWS. The proposed filesystem, named Fejk Filesystem (FFS), provides users with free, deniable, and cryptographic storage by exploiting the storage provided by online web services. The thesis analyzes and compares the performance of FFS against two other filesystems and a version of FFS that does not use an OWS. While FFS has performance limitations that make it non-viable as a general-purpose filesystem, such as a substitute for the local filesystem on a computer; however, FFS provides security benefits compared to other cloud-based filesystems specifically by providing end-to-end encryption, authenticated encryption, and plausible deniability of the data. Furthermore, being a cloud-based filesystem, FFS can be mounted on any computer with the same operating system, given the correct secrets. / Flertalet webbtjänster idag, så som Flickr och Twitter, tillhandahåller möjligheten att lägga upp bilder som lagras på tjänsten utan kostnad. Denna avhandling utforskar en ide om att skapa ett kryptografiskt säkert filsystem som lagrar sin data på webbtjänster med hjälp av kodade och krypterade bilder. Mer data kan ofta lagras i bildinlägg än i textinlägg på webbtjänsterna. Filsystemet, döpt Fejk Filesystem (FFS), tillhandahåller användare med gratis, förnekningsbar, och kryptografisk lagring genom att utnyttja det lagringsutrymme som tillhandahålls av dessa webbtjänster. Avhandlingen analyserar och jämför prestandan hos FFS med två andra filsystem, samt en version av FFS som inte använder en webbtjänst. En slutsats som kan dras är att FFS har begränsningar i prestanda, vilket gör det ohållbart som ett generellt-ändamålsfilssystem, till exempel som ersättare för det lokala filsystemet på en dator. Däremot tillhandahåller det säkerhetsförmåner jämfört med andra molnbaserade filsystem, så som början-till-slutetkryptering, autentiserad kryptering, samt sannolik förnekningsbarhet av datan. Dessutom, som ett molnbaserat filsystem kan FFS monteras på vilken dator som helst med samma operativsystem, givet de korrekta nycklarna.

Der Vortrag stellt die für den Anwender und Administrator sichtbaren Neuerungen im OpenAFS dar, die seit der Veröffentlichung der Version 1.0 bis zur Version 1.2.2 vorgenommen wurden.

HelenOS is an experimental operating system based on microkernel. It's developed by people related to Faculty of Mathematics and Physics at Charles University in Prague. The present implementantion is prepared to support a scale of filesystems because of VFS abstraction, but actually only a few of filesystems are supported. Ext4 is modern filesystem, developed Linux kernel developers. The previous releases ext2 and especially ext3 are still very popular in many Linux distributions. Imple- mentation of ext2/3/4 filesystems can be also found in other unix-based operating systems, for example *BSD or OpenSolaris. The main goal of this thesis is to extend HelenOS operating system to support previously mentioned filesystems.

Recently, Handheld devices are more and more popular. Most of them aim at the low price, small size, high computing power, and powerful functionalities. Therefore, the need for embedded operating systems in the market is absolutely vital. There are many embedded systems in the market, but embedded Linux has some advantages to be outstanding and widely accepted. For example, it has no proprietary problem and high portability, and is comparatively easy to be reconstructed and to develop new applications. Best of them all, it is open source software. Embedded systems are usually diskless systems. In order to keep permanent data in embedded Linux, using flash disk as its disk system is a widely adapted strategy. We use MTD (Memory Technology Devices) system to emulate flash memory as flash disk mounted into Linux virtual file system. This allows accessing flash memory with standard I/O operations without any extra effort. MTD system contains, ¡§user¡¨ and driver¡¨, two different modules. In driver modules, we use CFI (Common Flash Interface) to probe the flash chip and then partition it, while we use MTD BLOCK to emulate the flash partitions as block devices in user modules, and then mount them into Linux¡¦s virtual file system with JFFS2 (Journaling Flash File System version 2) type, which is a file system type specifically designed for flash devices according to the features of flash devices. The purpose of this thesis is to use MTD system to emulate Am29LV320DB flash chip as a flash disk in embedded Linux running on an ARM-based developing board, SMDK2410, designed by Samsung. I hope this porting can help the development of other advanced applications and provide an empirical platform for the research of embedded systems.

Workshop "Netz- und Service-Infrastrukturen" Der Vortrag beschreibt das Vorgehen bei der Umstellung des Authentication Services im URZ der TU Chemnitz. Seit 1994 wird der AFS kaserver betrieben und soll nun durch eine Kerberos V5 Installation ersetzt werden. Es erfolgt eine Analyse der Auswirkungen auf verschiedene Anwendungen und eine Auswahl zwischen den beiden verfügbaren Implementationen MIT Kerberos V5 und KTH Heimdal.

The master's thesis deals with a support of caching in GVfs (Gnome Virtual filesystem). Basics of caches, cache invalidation, and cache replacement algorithms are described. Description of GIO filesystem abstraction and communication of modules within GVfs is provided next. The disadvantages of individual GVfs modules and, if present, their internal caches are discussed. The thesis proposes three types of cache. The first type of cache is for storing file metadata, the second one for directory listings, and the latter for file content. These caches have been implemented in a prototype and verified with respect to the functionality and performance. The main benefits of the proposed solution are faster work with virtual filesystems and provided missing functionality of lower-level virtual filesystem to GIO abstraction (namely, for instance seek operation).

Der Vortrag stellt die für den Anwender und Administrator sichtbaren Neuerungen im OpenAFS dar, die seit der Veröffentlichung der Version 1.0 bis zur Version 1.2.2 vorgenommen wurden.

No description available.