无状态卸载包括以下功能: • Checksum Offload- 校验和卸载 • Large Send Offloads - 大发送包卸载 • Receive Side Scaling - 弹性接收 • Transmit Side Scaling - 弹性发送 • Interrupt Moderation - 中断自适应 • Large Receive Offloads - 大接收包卸载 • VLAN insertion and stripping - VLA...
Network Interface Controller Drivers 网卡的特性(speed, link status, LRO, LSO, RSS, Flow control, crc/checksum offload),各个网卡驱动(虚拟设备的驱动,I40e, ixgbe , Mlx5)等等 Release Notes 发布记录,每个版本的new features, bugfixes, known issues 等; FAQ dpdk中的常见的问题; dpdk支持的硬件信息 h...
tx-checksum-ipv4: off [fixed] tx-checksum-ip-generic: on tx-checksum-ipv6: off [fixed] tx-checksum-fcoe-crc: off [fixed] tx-checksum-sctp: off [fixed] scatter-gather: on tx-scatter-gather: on tx-scatter-gather-fraglist: off [fixed] tcp-segmentation-offload: on tx-tcp-segmentation: ...
• Checksum Offload- 校验和卸载 • Large Send Offloads - 大发送包卸载 • Receive Side Scaling - 弹性接收 • Transmit Side Scaling - 弹性发送 • Interrupt Moderation - 中断自适应 • Large Receive Offloads - 大接收包卸载 • VLAN insertion and stripping - VLAN 插入和剥离 ...
TCP/UDP/IP stateless offload LSO, LRO, checksum offload Receive Side Scaling (RSS) also on encapsulated packet Transmit Side Scaling (TSS) VLAN and MPLS tag insertion/stripping Receive flow steering See more Storage Accelerations NVMe over Fabric offloads for target Storage protocols...
It tunnels Ethernet frames within Ethernet + IP + UDP frames. Mellanox implements VXLAN encapsulation and decapsulation in the hardware. [Beta Level] Added support for NVMe over Fabrics (NVME-oF) offload, an implementation of the new NVME-oF standard target (server) side in hardware. [Beta ...
Related Documents High performance single/Dual port support with full line rate, full duplex FDR InfiniBand and up to 100GbE RDMA and RoCE (RDMA Over Converged Ethernet) Transport offload with zero copy for low CPU utilization TCP/IP stateless offload, and Hardware Checksum Offload for Tx and ...
让网卡会计算,还要让网卡承担安全、加密的智能,具备独立编程的能力。 按照这个思路, Mellanox在9月3日对外发布了ConnectX-6 Dx、BlueField-2两款芯片,对外称:新一代云Smart NIC和I/O 处理器 (IPU)解决方案。 根据介绍,ConnectX-6 Dx Smart NIC可提供两个 25 Gb/s、50 Gb/s或 100Gb/s 端口,或者单个 20...
(aka GPUDirect) communication acceleration –– Enhanced Atomic operations –– Registration-free RDMA memory access –– NVMe-over-Fabric hardware acceleration Transport Offloads – RDMA over Converged Ethernet (RoCE) –– TCP/UDP/IP stateless offload –– LSO, LRO, checksum offload –– RSS (...
Offloads1 TCP/UDP/IP stateless offload LSO,LRO, checksum offload RSS (also on encapsulated packet), TSS, HDS, VLAN and MPLS tag insertion/striping, Receive flow steering Data Plan Development Kit (DPDK) for kernel bypass applications Open VSwitch (OVS) offload using ...