Dell PowerEdge M610x - Manual (Page 28)

Dell
Dell PowerEdge M610x Technical Guide 23
10 Embedded NICs/LAN on Motherboard (LOM)
10.1 Overview
Two embedded Broadcom ® 5709S dual-port LAN controllers are on the M610x planar as independent
Gigabit Ethernet interface devices. The following information details the features of the LAN
devices:
•
x4 PCI Express Gen2 capable interface
• The M610 operates this controller at Gen1 speed
• Integrated MAC and PHY
•
3072x18 Byte context memory
•
64 KB receive buffer
• TOE (TCP Offload Engine)
•
NC-SI (Network Controller-Sideband Interface) connection for manageability
•
Wake-On-LAN (WOL)
• PXE 2.0 remote boot
•
iSCSI boot
•
IPv4 and IPv6 support
•
Bare metal deployment support
• ISCSI offload accelerator: used for offloading ISCSI traffic as an ISCSI accelerator/HBA –
Optionally enabled through a hardware key
The embedded NICs are not sharable with iDRAC since the blade iDRAC has a dedicated 100Mbps link
(Fabric D).
10.2 Platform Networking LAN on Motherboard (LOM) Technology
Overview
The PowerEdge M610x has two Ethernet ports because they include two built-in dual-port 1GbE
converged networking (CNIC) LOMs based on Broadcom 5709 controllers. The M610x supports multiple
functions over a unified fabric to help manage Ethernet, iSCSI and remote management traffic on
each port simultaneously.
Enterprise networks that use multiple protocols and multiple network fabrics benefit from the
Broadcom C-NICs LOMs’ ability to combine network traffic, storage, and clustering over a single
Ethernet fabric by boosting server CPU processing performance and memory utilization while
alleviating I/O bottlenecks.
Each BCM5709S LOM provides dual 10/100/1000BASE-T Gigabit Ethernet functions, an IEEE802.3-
compliant media access controller (MAC), and a UTP copper physical layer transceiver solution for
high-performance network applications. It enables simultaneous convergence of all networked
communications possible in a server, such as data network (LAN), storage network (e.g., block, iSCSI,
or file [e.g., CIFS/NFS]), and clustering (e.g., High-Performance Computing [HPC]).