This file was created by the TYPO3 extension bib --- Timezone: UTC Creation date: 2024-11-21 Creation time: 14-18-52 --- Number of references 9 inproceedings 2021-krude-nfp-pred Determination of Throughput Guarantees for Processor-based SmartNICs 2021 12 7 maki https://www.comsys.rwth-aachen.de/fileadmin/papers/2021/2021-krude-nfp-pred.pdf ACM The 17th International Conference on emerging Networking EXperiments and Technologies (CoNEXT '21) 978-1-4503-9098-9/21/12 10.1145/3485983.3494842 1 JohannesKrude JanRüth DanielSchemmel FelixRath Iohannes-HeorhFolbort KlausWehrle inproceedings 2019-krude-online-reprogramming Online Reprogrammable Multi Tenant Switches 2019 12 9 maki https://www.comsys.rwth-aachen.de/fileadmin/papers/2019/2019-krude-online-reprogramming.pdf ACM 1st ACM CoNEXT Workshop on Emerging in-Network Computing Paradigms (ENCP '19) 978-1-4503-7000-4/19/12 10.1145/3359993.3366643 1 JohannesKrude JacoHofmann MatthiasEichholz KlausWehrle AndreasKoch MiraMezini inproceedings 2019-glebke-in-network-cv Towards Executing Computer Vision Functionality on Programmable Network Devices 2019 12 9 reflexes,maki,internet-of-production https://www.comsys.rwth-aachen.de/fileadmin/papers/2019/2019-glebke-in-network-cv.pdf Online ACM 1st ACM CoNEXT Workshop on Emerging in-Network Computing Paradigms (ENCP '19) en 978-1-4503-7000-4/19/12 10.1145/3359993.3366646 1 RenéGlebke JohannesKrude IkeKunze JanRüth FelixSenger KlausWehrle inproceedings 2019-krude-chain-opt Optimizing Data Plane Programs for the Network 2019 8 23 With the move of Software-defined networking from fixed to programmable data planes, network functions are written with P4 or eBPF for targets such as programmable switches, CPU based flow processors and commodity CPUs. These data plane programs are, however, limited in per-packet time budget (e.g., 67.2 ns at 10GbE) and program size, making program optimization imperative. Existing approaches focus on optimizing the distribution of flow rules in fixed data planes or they are limited to a single switch. We see great potential in integrating the network topology into program optimization. maki https://www.comsys.rwth-aachen.de/fileadmin/papers/2019/2019-krude-chain-opt.pdf ACM NetPL '19: ACM SIGCOMM Workshop on Networking and Programming Languages Beijing, China 978-1-4503-6877-3/19/08 10.1145/3341561.3349590 1 JohannesKrude MatthiasEichholz MaximilianWinck KlausWehrle MiraMezini inproceedings 2019-hohlfeld-bpfperf Demystifying the Performance of XDP BPF 2019 6 25 maki,reflexes https://www.comsys.rwth-aachen.de/fileadmin/papers/2019/2019-hohlfeld-bpfperf.pdf IEEE IEEE International Conference on Network Softwarization (NetSoft) IEEE International Conference on Network Softwarization 10.1109/NETSOFT.2019.8806651 1 OliverHohlfeld JohannesKrude Jens HelgeReelfs JanRüth KlausWehrle inproceedings 2018-krude-circuit Circuit Switched VM Networks for Zero-Copy IO 2018 8 20 1-7 maki https://www.comsys.rwth-aachen.de/fileadmin/papers/2018/2018-krude-xocks.pdf ACM Proceedings of the 2018 Afternoon Workshop on Kernel Bypassing Networks (KBNets'18) Budapest, Hungary Afternoon Workshop on Kernel Bypassing Networks 20.8.2018 10.1145/3229538.3229539 1 JohannesKrude MirkoStoffers KlausWehrle inproceedings 2017-teubert-secuware-hugin Hugin: A Scalable Hybrid Android Malware Detection System 2017 9 168-176 Mobile operating systems are a prime target of today’s malware authors and cyber criminals. In particular, Google’s Android suffers from an ever increasing number of malware attacks in the form of malicious apps. These typically originate from poorly policed third-party app stores that fail to vet the apps prior to publication. In this paper, we present Hugin, a machine learning-based app vetting system that uses features derived from dynamic, as well as static analysis and thus falls into the scarcely studied class of hybrid approaches. Hugin is unique with respect to using IPC/RPC monitoring as source for dynamically extracted features. Furthermore, Hugin uses a short (and yet effective) feature vector that leads to a high efficiency in training as well as classification. Our evaluation shows that Hugin achieves a detection accuracy of up to 99.74% on an up-to- date data set consisting of more than 14,000 malware samples and thus, is easily capable of competing with other current systems. mobile malware detection; app vetting; machine learning http://www.thinkmind.org/download.php?articleid=securware_2017_10_20_30067 IARIA SECUWARE 2017 The Eleventh International Conference on Emerging Security Information, Systems and Technologies Hugin: A Scalable Hybrid Android Malware Detection Syste September 2017 978-1-61208-582-1 1 1 DominikTeubert JohannesKrude SamuelSchüppen UlrikeMeyer proceedings 2017-SymPerfPoster SymPerf: Predicting Network Function Performance 2017 8 21 spp,erc,symbiosys,reflexes https://www.comsys.rwth-aachen.de/fileadmin/papers/2017/2017-rath-sym-perf-poster.pdf ACM Los Angeles, USA ACM SIGCOMM 2017 Poster 21.8.2017 - 25.8.2017 en 978-1-4503-5057-0/17/08 10.1145/3123878.3131977 1 FelixRath JohannesKrude JanRüth DanielSchemmel OliverHohlfeld Jó AgilaBitsch Link KlausWehrle inproceedings 2013-krude-ccsw-sandbox A Versatile Code Execution Isolation Framework with Security First 2013 11 1-10 In PaaS architectures, code execution needs to be isolated to protect tenants from unauthorized access to their data by other tenants and to protect the host system from any type of intrusion by other tenants. In this paper we propose a simple and lightweight framework that provides such code execution isolation units. All design decisions taken for our framework follow well-established design principles for protection mechanisms, such that in our design we put security first. Our framework uses the process barrier and the seccomp filter mechanism to restrict access to memory and to the system call interface. For access to additional resources a request delegation scheme is used. We evaluate our implementation on a small PaaS example architecture. For plain computations, the applied techniques do not introduce any overhead. In general, setup time for isolation units is negligible and the framework achieves acceptable throughput for broad network access. Our work brings tight and versatile isolation units to current unmodified Linux operating systems. ACM Proceedings of the 2013 ACM workshop on Cloud computing security workshop (CCSW '13) 978-1-4503-2490-8 10.1145/2517488.2517499 1 1 JohannesKrude UlrikeMeyer