Features
The Intel X550 adapter has the following features:
Supports Intel Virtualization Technology for connectivity (VT-c), I/O virtualization advances network
connectivity used in today’s servers to more efficient models by providing Flexible Port Partitioning
(FPP), multiple Tx/Rx queues, Tx queue rate-limiting, and on-controller QoS functionality that is
useful for both virtual and non-virtual server deployments.
Supports Virtual Machine Device Queues (VMDq) for NIC-based VM queue sorting, enabling efficient
hypervisor-based switching. VMDq reduces I/O impact on the hypervisor in a virtualized server by
performing data sorting and coalescing in the network adapter.
Supports SR-IOV for direct assignment - NIC-based isolation and switching for various virtual
station instances enabling optimal CPU usage in virtualized environment.
Provides virtual bridging support that delivers both host-side and switch-side control and
management of virtualized I/O as well as the following modes of virtualized operation:
VEPA: IEEE 802.1Qbg support for Virtual Ethernet Port Aggregator.
VEB: Virtual Ethernet Bridge support with Intel VT.
Supports VXLAN/NVGRE Hardware Offloads, stateless offloads that preserve application
performance for overlay networks. With these offloads, it is possible to distribute network traffic
across CPU cores. At the same time, the X550 offloads LSO, GSO, and checksum from the host
software, which reduces CPU overhead.
Flexible Port Partitioning (FPP), based on the SR-IOV specification, enables virtual Ethernet
controllers that can be used by a Linux host directly or assigned directly to virtual machines for
hypervisor virtual switch bypass. FPP enables the assignment of up to 64 Linux host processes or
virtual machines per port to virtual functions. FPP can be used to control the partitioning of the
bandwidth across multiple virtual functions. FPP can also provide balanced QoS by giving each
assigned virtual function equal access to 10 Gb/s of bandwidth.
MSI-X interrupts support minimizes the impact of I/O interrupts by load balancing interrupts across
multiple processor cores.
Low-Latency Interrupts: Allows the adapter to bypass the automatic moderation of time intervals
between the interrupts (based on the sensitivity of the incoming data).
Load balancing on multiple processors, which increases performance on multiprocessor systems by
efficiently balancing network loads across processor cores when used with Receive Side Scaling
(RSS) from Microsoft or Scalable I/O on Linux.
Header Splits and Replication in Receive helps the driver focus on the relevant part of the packet
without needing to parse it.
Multiple queues allow packet handling without the waiting/buffer overflow, which provides efficient
packet prioritization.
Mobile and cloud application acceleration: Intel’s Data Plan Development Kit (DPDK) delivers an
open driver where users can fine-tune small packet performance.
Offload features:
IP, TCP, and UDP checksum offload (IPv4 and IPv6)
TCP and UDP segmentation/large send offload (IPv4 and IPv6)
IPsec offload
Receive Side Scaling for Windows and Scalable I/O for Linux (IPv4, IPv6, and TCP/ UDP)
IEEE 802.1Q VLAN support with VLAN tag insertion, with stripping and packet filtering for up to
4096 VLAN tags.
IEEE 802.3x flow control support.
IEEE 802.1p Class of Service/Quality of Service.
Support for Advanced Packet Filtering.