Datasheet

Data Sheet
© 2011 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. Page 8 of 22
Cisco unified fabric consolidates all data center I/O onto Layer 2 Ethernet. Unified fabric reduces capital and
operating costs by reducing the number of server adapters, cables, and upstream switches needed. All I/O
(LAN, SAN, and cluster) typically is consolidated onto two Ethernet links. DCB and FCoE enable the
incorporation of Fibre Channel frames into a unified fabric, facilitating wire-once strategies in which all servers
become capable of SAN connection. A standard and uniform approach to I/O enhances server and storage
consolidation strategies. The Cisco Nexus 5500 platform also connects to existing native Fibre Channel
networks, protecting current investments in storage networks. Additionally, the Cisco Nexus 5500 platform
attaches to directly connected FCoE and Fibre Channel storage devices and supports multi-tiered unified
network fabric connect directly over FCoE.
Unified ports allow any capable port to take on the character of 1 and 10 Gigabit Ethernet, SAN and LAN
shared on 10 Gigabit Ethernet, or 8/4/2/1-Gbps Fibre Channel. Unified ports give the user flexibility in
choosing SAN and LAN port options consistent with the virtualized data center and offer a migration path to
FCoE for those users not yet ready to make the move from native Fibre Channel.
Energy efficiency achieved through the use of the Cisco Nexus 5500 platform helps data centers operate
within their space, power, and cooling parameters while reducing their carbon footprints. Every network link at
the rack level requires adapters, switches, and transceivers, all of which consume power. I/O consolidation
reduces energy consumption by eliminating the need for separate Fibre Channel adapters, cables, and
switches. In many cases, server cluster networks also can be consolidated onto 10 Gigabit Ethernet
networks, especially given the low latency of the Cisco Nexus 5500 platform. The switch hardware is also
designed for energy efficiency. Variable-speed fans consume only the amount of power necessary to cool the
chassis at that specific point in time. The switch power supplies are sized to support worst-case scenarios, in
which inefficient SFP+ transceivers increase power draw; however, when low-power cabling solutions are
deployed, the switch platform’s power supplies maintain 90 percent efficiency at only 25 percent utilization,
making efficient use of power in best-case scenarios.
Consistent management for Cisco products is provided through the consistency of both the Cisco NX-OS
Software and Cisco MDS 9000 Software management models and tools. The switch platform’s network
features can be managed using the Cisco command-line interface (CLI), and the Fibre Channel and FCoE
features can be managed through the Cisco Fabric Manager suite. Cisco Data Center Network Manager
(DCNM) also supports the Cisco Nexus 5500 platform. The capability to manage Ethernet and FCoE features
independently with existing Cisco tools preserves existing management models, best practices, and
investments in staff training. In addition, Simple Network Management Protocol (SNMP) MIBs, XML, and the
Cisco CLI are made available to customers for switch management through third-party and custom-
developed tools. The switch platform uses Cisco NX-OS for superior operating efficiency, pervasive security,
and continuous operation even through software upgrades.
Software manageability and serviceability features include Smart Call Home and automated parameter
exchange (through DCBX). Security is enhanced through role-based access control (RBAC); support for
authentication, authorization, and accounting (AAA), remote TACACS+, and RADIUS servers; and Secure
Shell (SSH) access.
Applications
The Cisco Nexus 5500 platform supports a number of application scenarios, making the switches a versatile data
center option:
As access-layer switches, they can be used purely as 1 and 10 Gigabit Ethernet switches, consolidating 10
Gigabit Ethernet connections into a smaller number of server connections trunked to the aggregation layer.