ProLiant cClass NC364m (Intel) Quad Port PCI-E Gigabit Server Adapter

Description

Quad-port Gigabit Ethernet throughput

Up to 2000 Mbps full duplex Ethernet transfer rate per port (8000 Mbps combined) delivers outstanding network performance designed to improve response time and remove bottlenecks across the entire network. Four ports on a single adapter are ideal for high density; I/O slot constrained blade servers, virtualization and security, applications, server consolidation, and increased network segmentation. The total aggregate throughput of 4Gbps per direction also meets the needs for customers desiring high bandwidth but are not yet ready to move to 10 Gigabit Ethernet.

Jumbo frames

Jumbo frames (also known as extended frames) permit a 9K byte transmission unit (MTU), which is six times the size of standard 1500 byte Ethernet frame. The NC364m supports jumbo frames as a way to achieve higher throughput and better CPU utilization. Jumbo frames are particularly useful for database transfers and tape backups.

802.1Q VLANs with 802.1p QoS tagging

IEEE 802.1Q virtual local area network (VLAN) protocol allows each physical port of the NC364m to be separated into multiple virtual NICs for added network segmentation and enhanced security and performance. VLANs increase security by isolating traffic between users. Limiting the broadcast traffic to within the same VLAN domain also improves performance. IEEE quality of service (QoS) 802.1p tagging allows the adapter to mark or tag frames with a priority level across a QoS-aware network for improved traffic flow.

TCP/IP stateless offloading

For overall improved system response, the NC364m supports standard TCP/IP offloading techniques including:TCP/IP checksum offload (TCO) moves the TCP and IP checksum offloading from the CPU to the network adapter. Large send offload (LSO) or TCP segmentation offload (TSO) allows the TCP segmentation to be handled by the adapter rather than the CPU.

Interrupt coalescence

Interrupt coalescing (interrupt moderation) groups multiple packets, thereby reducing the number of interrupts sent to the host. This process optimizes host efficiency, leaving the CPU available for other duties.

Dual address cycles

Dual address cycles (DAC) provide the ability to address memory above 4 GB and improve system performance by preventing the server operating system from performing a buffer copy from below 4 GB to above 4 GB.