National Repository of Grey Literature 33 records found  1 - 10nextend  jump to record: Search took 0.02 seconds. 
Acceleration of Open vSwitch
Vodák, David ; Orsák, Michal (referee) ; Martínek, Tomáš (advisor)
Virtual switch is a program, which is used for connecting virtual machines to network and that is why it is a crucial part of server virtualization. However virtual switch is consuming too much performance of the server which it is running on. A measurement of Open vSwitch (OvS) indicates that for data speed of 10 Gb/s, approximately 4 cores of the processor are fully occupied. As the consumption of performance is directly proportional to transmission speed, it may eventually get to the point where the consumption of performance cannot be handled. This bachelor thesis is about acceleration of the Open vSwitch with the help of the DPDK Poll Mode Driver extended by support of the SR-IOV virtualization technology as well as the interface for offloading classification rules to hardware called RTE flow. In the scope of this thesis the SR-IOV is implemented and then tested on OvS. Furthermore, the RTE flow support was designed and partially implemented.
System for Monitoring of Network Protocols
Selecký, Roman ; Dražil, Jan (referee) ; Kořenek, Jan (advisor)
It is necessary to monitor networks namely for diagnostics, troubleshooting, detection of anomalies and suspicious header encapsulations. This thesis aims to design and implement a system for monitoring protocol structure on 10 Gb networks, which will be able to capture packets based on the sequence of encapsulated protocols. To achieve requested throughput some tasks like packet parsing and packet filtering were accelerated in FPGA. Flexibility is achieved by using a tool that maps P4 programs, which define packet parsing process, to VHDL language. Based on the information gained from packet parsing, flow records are created and stored via IPFIX protocol. This information is displayed through a graphical user interface in the form of protocol tree, whose nodes are associated with flow records.
Protection Against DoS Attacks Using P4 Language
Vojanec, Kamil ; Fukač, Tomáš (referee) ; Kučera, Jan (advisor)
This thesis focuses on reimplementation of existing DoS (Denial of Service) attack mitigation device with high-level P4 programming language. The main reason for using P4 is to enhance adaptability and functionality to different types of DoS attacks. The created device is designed in a modular way and enables easy alterations by using interchangeable components. The target platform for this thesis is an FPGA acceleration card. The work results in designing several DoS mitigation components and implementing applications composed of these components. Pats of this work have been presented at IEEE ANCS (Symposium on Architectures for Networking and Communication Systems) in September 2019 at University of Cambridge.
Mapping of Match Tables from P4 Language to FPGA Technology
Kekely, Michal ; Matoušek, Jiří (referee) ; Kořenek, Jan (advisor)
This thesis deals with design and implementation of mapping of match action tables from P4 language to FPGA technology. Goal of the thesis was to describe key principles, which need to be understood in order to design such a mapping and function of algorithms needed, apply these principles by implementing them and analyze the speed and memory requirements of such an implementation. Outcome provides configurable hardware unit capable of classifying packets and connection between the unit and match action tables from P4 language. The implementation is based on DCFL algorithm and requires less memory compared to HiCuts and HyperCuts algorithms while being comparably fast at worst-case scenarios.
Remote Configuration of P4 Device
Neruda, Jakub ; Kučera, Jan (referee) ; Wrona, Jan (advisor)
Administration of a large network from a central node with a vendor independent API is quite important issue these days. The concept of SDN was truly helpful with realization of the solution, namely in the form of the OpenFlow protocol. Nowadays, a P4 language is gaining momentum, primarily thanks to its ability to describe whole packet processing pipeline and also for the P4 Runtime, solution to the distributed network configuration. CESNET association is one of the research groups starting to support P4 in their network cards beloging to the Combo series. In this work, an API was designed for these cards, aimed at the dynamic flow table configuration. This API was used for implementation of a basic support of the Combo cards in the P4 Runtime.
Network Traffic Monitoring Using the P4 Language
Patová, Pavlína ; Matoušek, Jiří (referee) ; Martínek, Tomáš (advisor)
Today we often encounter the need to monitor network and service quality. For this purpose we can use for example INT. Our goal is to find the optimal platform and associated compiler for implementing INT. We will try to find an alternative to the existing solutions (T4P4S, BMv2). However, we will also mention these two platforms and discuss their advantages and disadvantages. The result of this work is an overview of the capabilities of each compiler and the performance of the described implementations.
OVS Acceleration Using FPGA Acceleration Card
Vido, Matej ; Martínek, Tomáš (referee) ; Kořenek, Jan (advisor)
The performance of the virtual switch Open vSwitch (OVS) is insufficient to satisfy the current requirements for link bandwidth of the server connections. There is an effort to accelerate the OVS both in the software and in the hardware by offloading the datapath to the smart network interface cards. In this work the COMBO card for 100G Ethernet developed by CESNET is used to accelerate the OVS. The suggested solution utilizes the firmware for FPGA generated from the definition in the P4 language to classify the packets in the card and DPDK for the data transfers and offloading the classification rules into the card. Forwarding of one flow with the shortest frames from physical to physical interface using one CPU core reaches forwarding rate of 11.2 Mp/s (10 times more than the standard OVS) with classification in the card and 5.9 Mp/s without classification in the card.
Flexible Load Balancer Using P4 Language
Šesták, Jindřich ; Fukač, Tomáš (referee) ; Martínek, Tomáš (advisor)
Currently servers of internet services are usually grouped together into clusters to provide sufficient performance to serve clients' queries. Each cluster needs Load Balancer, so it can choose one server which will process query from one client. For describing such device that processes packets is convenient to use P4 language. Within this work, the principles of load balancing, design, implementation and testing of a simple Load Balancer described in P4 language were demonstrated. The program is tested using Behavioral model of P4 language on a common processor and on the NFB-200G2QL card thanks to the Netcope environment from the CESNET association
High level synthesis in network applications described using P4 language
Panák, Petr ; Šťáva, Martin (referee) ; Fujcik, Lukáš (advisor)
High-level synthesis is a compelling method of designing a digital circuit. High abstraction and faster verification are advantages which aren't pressent in Register Transfer Level designing. That guarantees faster designing with lower development costs. This bachelor thesis deals with a digital design of actions, extern blocks and MI32 interface access. Each component design is described using C/C++ programming language and synthesised with Intel HLS compiler.
The Stateful Packet Processing in P4 Language
Kohout, Pavel ; Martínek, Tomáš (referee) ; Kořenek, Jan (advisor)
With the growing speed and complexity of computer networks, arise requirements for creating powerful devices that are capable of collecting statistics and changing their own functionality according to the demands of network administrators. These requirements can be described using specialized programming languages such as P4. In this bachelor thesis a design, implementation, testing and integration of register and counter stateful memory modules into P4 compiler system for FPGA technology was made. The created system supports the collection of statistics described in P4 language at speeds up to 100 Gbps.

National Repository of Grey Literature : 33 records found   1 - 10nextend  jump to record:
Interested in being notified about new results for this query?
Subscribe to the RSS feed.