This thesis is based on seven publications related to the area of digital forensics which were published at conferences or in journals by ACM, IEEE, USENIX and IFIP. Digital forensics as research field has received increasing attention in recent years, as more and more crimes are committed exclusively or with the involvement of computers. At the same time, new challenges emerge constantly, e.g. the prevalent use of encryption, mobile devices of various nature, online cloud storage services and readily available tools that facilitate counter-forensics. In particular, this thesis tries to mitigate current challenges for digital forensics in the areas of online data storage and anonymous communication. Regarding anonymous communication we analyzed the well-known online anonymity tool Tor, which employs onion routing and is expected to be used by hundreds of thousands of users every day: firstly how it is used, and secondly what can be learnt from the publicly available server information. We were able to show that the majority of users are not employing Tor as recommended by the Tor community, and we found many information leaks that can endanger the users- anonymity. We also studied how the underlying infrastructure, which is run by volunteers, can be monitored to provide useful metrics of interest. We furthermore derived and implemented a new attack on online storage systems abusing client-side data deduplication and analyzed how it can be used to thwart digital forensic investigations which in turn can be used for forensic investigations. We showed its feasibility on Dropbox, one of the largest cloud storage providers with more than 200 million users worldwide at the time of writing this thesis. We quantified slack space on numerousWindows systems, assessed it-s stability over time regarding system updates and found that up to 100 megabytes of slack space are readily available in files of the operating system. We furthermore implemented a digital alibi framework with a social interaction component which in our opinion can be easily overlooked in forensic analysis as conducted today. Finally we analyzed browser artifacts and how they can be used for browser fingerprinting. We then used browser fingerprinting to enhance HTTP session security by binding the session on the server to specifics of the particular browser used.