Vše

Co hledáte?

Vše
Projekty
Výsledky výzkumu
Subjekty

Rychlé hledání

  • Projekty podpořené TA ČR
  • Významné projekty
  • Projekty s nejvyšší státní podporou
  • Aktuálně běžící projekty

Chytré vyhledávání

  • Takto najdu konkrétní +slovo
  • Takto z výsledků -slovo zcela vynechám
  • “Takto můžu najít celou frázi”

400G Packet Parsing on an FPGA-Based SmartNIC

Identifikátory výsledku

  • Kód výsledku v IS VaVaI

    <a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F63839172%3A_____%2F23%3A10133581" target="_blank" >RIV/63839172:_____/23:10133581 - isvavai.cz</a>

  • Výsledek na webu

  • DOI - Digital Object Identifier

Alternativní jazyky

  • Jazyk výsledku

    angličtina

  • Název v původním jazyce

    400G Packet Parsing on an FPGA-Based SmartNIC

  • Popis výsledku v původním jazyce

    CESNET, the NREN (National Research and Education Network) of the Czech Republic, is ready to present its newest research in the area of networking technologies. This includes a 400G SmartNIC, an FPGA-based application built on the NDK (Network Development Kit) framework, and a Linux driver. The NDK is open-source and available on GitHub. It was designed specifically for high-speed packet processing, which is possible due to the MFB and MVB buses that allow the processing of multiple packets in each clock cycle. The proposed demo will showcase packet parsing at 400 Gbps on the Reflex CES XpressSX AGI-FH400G board with an Intel Agilex I-Series FPGA. The whole FPGA design for the demo is built on the NDK, although the Parser unit and the DMA Module6 are not a part of the GitHub repository. The Parser can parse most protocols encapsulated in Ethernet frames, namely VLAN(s), MPLS(s), IPv4, IPv6, ICMP, TCP, UDP, and SCTP. During the demo, packets (stored in a PCAP) will be loaded into the PCAP Player in the FPGA. From there, they will be replayed, going out of the FPGA toward the card&apos;s transceiver. The transceiver with a loopback module will return the packets to the FPGA, where the throughput measuring and protocol parsing will occur. The Parser will extract protocol header fields and send them to counters that will be read once the replaying stops. Unparsed packets are also gradually received back into the host PC (via the 400G DMA Module) and stored in a PCAP to be compared with the input PCAP. Ultimately, the throughput will be calculated from the number of transferred bytes in a certain amount of time obtained from the FPGA design.

  • Název v anglickém jazyce

    400G Packet Parsing on an FPGA-Based SmartNIC

  • Popis výsledku anglicky

    CESNET, the NREN (National Research and Education Network) of the Czech Republic, is ready to present its newest research in the area of networking technologies. This includes a 400G SmartNIC, an FPGA-based application built on the NDK (Network Development Kit) framework, and a Linux driver. The NDK is open-source and available on GitHub. It was designed specifically for high-speed packet processing, which is possible due to the MFB and MVB buses that allow the processing of multiple packets in each clock cycle. The proposed demo will showcase packet parsing at 400 Gbps on the Reflex CES XpressSX AGI-FH400G board with an Intel Agilex I-Series FPGA. The whole FPGA design for the demo is built on the NDK, although the Parser unit and the DMA Module6 are not a part of the GitHub repository. The Parser can parse most protocols encapsulated in Ethernet frames, namely VLAN(s), MPLS(s), IPv4, IPv6, ICMP, TCP, UDP, and SCTP. During the demo, packets (stored in a PCAP) will be loaded into the PCAP Player in the FPGA. From there, they will be replayed, going out of the FPGA toward the card&apos;s transceiver. The transceiver with a loopback module will return the packets to the FPGA, where the throughput measuring and protocol parsing will occur. The Parser will extract protocol header fields and send them to counters that will be read once the replaying stops. Unparsed packets are also gradually received back into the host PC (via the 400G DMA Module) and stored in a PCAP to be compared with the input PCAP. Ultimately, the throughput will be calculated from the number of transferred bytes in a certain amount of time obtained from the FPGA design.

Klasifikace

  • Druh

    O - Ostatní výsledky

  • CEP obor

  • OECD FORD obor

    20206 - Computer hardware and architecture

Návaznosti výsledku

  • Projekt

    Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

  • Návaznosti

    P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Ostatní

  • Rok uplatnění

    2023

  • Kód důvěrnosti údajů

    S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů