Data Sheet
Cisco Nexus 5600 Platform 10-Gbps Switches
Product Overview
In today’s data centers, virtualization deployments have become commonplace, and this trend is increasing rapidly with the availability of 10 Gigabit Ethernet servers at attractive prices. The combination of increased adoption of these servers and applications with higher bandwidth requirements is increasing the need for dense 10 and 40 Gigabit Ethernet switching. Moreover, data center architectures are changing as application environments create new demands for IT infrastructure. Application workloads are deployed across a mix of virtualized and nonvirtualized server and storage infrastructure, requiring a network infrastructure that provides consistent connectivity, security, and visibility across a range of bare-metal, virtualized, and cloud computing environments.
The Cisco Nexus® 5600 platform is the third generation of the Cisco Nexus 5000 Series Switches: the industry’s leading data center server access switches. The Cisco Nexus 5600 platform switches can be categorized into 10-Gbps and 40-Gbps switches. This data sheet focuses on the 10-Gbps switches only. Cisco Nexus 5600 platform 10-Gbps switches are the successors to the industry’s widely adopted Cisco Nexus 5500 platform switches. The switches maintain all the existing Cisco Nexus 5500 platform features, including LAN and SAN convergence (unified ports and Fibre Channel over Ethernet [FCoE]), fabric extenders, and Cisco® FabricPath. In addition, the Cisco Nexus 5600 platform 10-Gbps switches bring integrated line-rate Layer 2 and 3 capabilities with true 40 Gigabit Ethernet support (on uplink and network-facing ports), Cisco programmable fabric innovations, Network Virtualization Using Generic Routing Encapsulation (NVGRE), Virtual Extensible LAN (VXLAN) bridging and routing, network programmability and visibility, large buffer capacity, and significantly greater scalability and performance for highly virtualized, automated, and cloud environments.
The Cisco Nexus 5600 platform 10-Gbps switches include both 1-rack-unit (1RU) and 2RU switches built to meet the challenges of today’s data centers with a flexible, agile, and energy-efficient design. These 10-Gbps switches are an important component of the Cisco Unified Data Center architecture, complementing existing Cisco Nexus switches. These energy-efficient switches offer 10 and 40 Gigabit Ethernet and FCoE, providing integrated Layer 2 and 3 features at wire speed and low latency of approximately 1 microsecond for any packet size. With a choice of port-side intake and fan-side intake airflow options to align with cold-aisle and hot-aisle placement in the data center, the 10-Gbps switches are designed for a broad range of traditional data center and large-scale virtualized cloud deployments.
The 10-Gbps switches together with the Cisco NX-OS Software operating system provides customers with features and capabilities that are widely deployed in data centers around the world. NX-OS is a purpose-built data center operating system designed for performance, resiliency, scalability, manageability, and programmability. It meets Ethernet and storage networking requirements, providing a robust and comprehensive feature set that can meet the demanding requirements of virtualization and automation in present and future data centers.
The Cisco Nexus 5600 platform 10-Gbps switches are designed for top-of-rack (ToR) and middle-of-row (MoR) deployment in data centers that support enterprise applications, service provider hosting, and cloud computing environments.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 1 of 24 |
Models and Configurations
The Cisco 5600 platform 10-Gbps switches come in these configurations.
Cisco Nexus 5672UP Switch
●The Cisco Nexus 5672UP Switch (Figure 1) is a 10 and 40 Gigabit Ethernet (40-Gbps on uplink and network-facing ports) switch offering wire-speed performance for up to seventy-two 10 Gigabit Ethernet ports (using Quad Small Form-Factor Pluggable [QSFP] breakout cables). The Cisco Nexus 5672UP Switches are Layer 2 and 3 nonblocking 10 and 40 Gigabit Ethernet and FCoE-capable switches with up to 1.44 terabits per second (Tbps) of internal bandwidth. The Cisco Nexus 5672UP offers 48 fixed 1 and 10 Gigabit Ethernet ports, of which the last 16 ports (highlighted in orange on the chassis for easy identification) are unified ports. All 48 fixed ports support classical Ethernet and FCoE. In addition, the 16 unified ports provide 8-, 4-, and 2-Gbps Fibre Channel, as well as 10 Gigabit Ethernet and FCoE connectivity options. The Cisco Nexus 5672UP also offers 6 ports of 40 Gbps using QSFP transceivers for Ethernet and FCoE support. The Cisco Nexus 5672UP has three fan modules and two power supplies. The Cisco Nexus 5672UP supports VXLAN in bridging and routing modes on all ports at line rate, enabling the migration of virtual machines between servers across Layer 3 networks. The switch also offers up to 15 buffer-to-buffer credits per port, helping provide SAN extension of up to 3.7 kilometers at 8-Gbps Fibre Channel speed assuming full Fibre Channel frame size of 2112 bytes.
Figure 1. Cisco Nexus 5672UP Switch (Port-Side View)
The Cisco Nexus 5672UP is constructed with the components shown in Figure 2. The Cisco Nexus 5672UP has two 1+1 redundant, hot-swappable power supplies and three hot-swappable independent fans with support for 2+1 redundancy.
Figure 2. Cisco Nexus 5672UP Switch (Fan-Side View)
The Cisco Nexus 5672UP supports both port-side intake (red handle) and fan-side intake (blue handle) airflow options for flexible mounting.
Cisco Nexus 5672UP-16G Switch
●The Cisco Nexus 5672UP-16G Switch (Figure 3) is a 10 and 40 Gigabit Ethernet (40-Gbps on uplink and network-facing ports) switch offering wire-speed performance for up to seventy-two 10 Gigabit Ethernet ports (using QSFP breakout cables). The Cisco Nexus 5672UP-16G Switches are Layer 2 and 3 nonblocking 10 and 40 Gigabit Ethernet and FCoE-capable switches with up to 1.44 Tbps of internal bandwidth.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 2 of 24 |
The Cisco Nexus 5672UP-16G offers 48 10 Gigabit Ethernet ports, of which the first 24 ports are capable of doing 1/10G Ethernet and the last 24 ports (highlighted in orange on the chassis for easy identification) are unified ports that support 16-, 8-, 4-, and 2-Gbps Fibre Channel. All 48 fixed ports support classical Ethernet and FCoE.
The 24 unified ports provide 16-, 8-. 4-, and 2-Gbps Fibre Channel as well as 10 Gigabit Ethernet and FCoE connectivity options. The Cisco Nexus 5672UP-16G also offers 6 ports of 40 Gbps using QSFP transceivers for Ethernet and FCoE support. The Cisco Nexus 5672UP-16G has three fan modules and two power supplies and supports VXLAN in bridging and routing modes on all ports at line rate, enabling the migration of virtual machines between servers across Layer 3 networks. The switch also offers up to 128 buffer-to- buffer credits per port, helping provide SAN extension of up to 16 kilometers at 16-Gbps Fibre Channel speed assuming full Fibre Channel frame size of 2122 bytes.
Figure 3. Cisco Nexus 5672UP-16G Switch (Port-Side View)
The Cisco Nexus 5672UP-16G platform is constructed with the components shown in Figure 4. The Cisco Nexus 5672UP-16G has two 1+1 redundant, hot-swappable power supplies and three hot-swappable independent fans with support for 2+1 redundancy.
Figure 4. Cisco Nexus 5672UP-16G Switch (Fan-Side View)
The Cisco Nexus 5672UP-16G supports both port-side intake (red handle) and fan-side intake (blue handle) airflow options for flexible mounting.
Cisco Nexus 56128P Switch
The Cisco Nexus 56128P Switch (Figure 5) is a 2RU switch that supports 2.56 Tbps of bandwidth across 48 fixed 1 and 10 Gigabit Ethernet SFP+ ports and four 40-Gbps QSFP+ ports. The 48 ports on the base chassis support 10 Gigabit Ethernet and FCoE. The 4 QSFP ports support 40 Gigabit Ethernet and FCoE.
Figure 5. Cisco Nexus 56128P Switch (Port-Side View)
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 3 of 24 |
The Cisco Nexus 56128P also offers two slots for a generic expansion module (GEM). The GEM (Figure 6) for the Cisco Nexus 56128P provides 24 SFP+ ports for 10 Gigabit Ethernet and FCoE or 2-, 4-, and 8-Gbps Fibre Channel, and 2 QSFP+ ports for 40 Gigabit Ethernet and FCoE. The expansion module supports native 40 Gigabit Ethernet on the QSFP+ ports. The GEM is supported on the Cisco Nexus 56128P chassis only and can be populated in either of the two expansion slots.
Figure 6. Cisco Nexus 56128P Generic Expansion Module
The Cisco Nexus 56128P is constructed with the components shown in Figure 7. The Cisco Nexus 56128P has four 2+2 redundant, hot-swappable power supplies and four 3+1 redundant, hot-swappable independent fans. The Cisco Nexus 56128P supports both port-side intake (red handle) and fan-side intake (blue handle) airflow options.
Figure 7. Cisco Nexus 56128P Switch (Fan-Side View)
The Cisco Nexus 56128P supports VXLAN bridging and routing modes on all ports at line rate, enabling the migration of virtual machines between servers across Layer 3 networks.
With the Cisco Nexus 5600 10-Gbps platform, organizations can quickly and easily upgrade existing data centers through advanced Cisco bidirectional (BiDi) optics, which enable the use of existing 10 Gigabit Ethernet fiber (a pair of multimode fiber strands) to carry 40 Gigabit Ethernet to the aggregation layer or to the spine (in a leaf-and- spine configuration) without requiring any change to the existing cabling infrastructure. Additionally, the platform can be deployed in MoR or EoR configurations to meet the 10 and 40 Gigabit Ethernet connectivity requirements of multiple racks or pods.
When used with Cisco Nexus 2000 Series Fabric Extenders, the Cisco Nexus 5600 platform 10-Gbps switches can support even more servers in a collapsed accessand aggregation-layer design, supporting 1 and 10 Gigabit Ethernet connectivity across multiple racks.
Features and Benefits
The following are some of the primary features of the Cisco Nexus 5600 10-Gbps platform switches:
●Optimization for virtualization and cloud deployments: Today, high-performance servers deployed in the cloud can support many more virtual machines and workloads than ever before. The requirement to be able to deploy new servers on demand puts additional strain on the network fabric. The 10-Gbps switches address this challenge by providing scalability and performance, making it an excellent platform for meeting current and future needs.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 4 of 24 |
●Density and resilience: Built for today’s data centers, the switches are designed just like the servers they support. Ports and power connections are at the rear, close to server ports, helping keep cable lengths as short as possible and delivering to rack servers benefits traditionally offered only on blade servers. Hotswappable power and fan modules can be accessed from the front panel, where status lights offer an at-a- glance view of switch operation. Front-to-back or back-to-front cooling is consistent with server designs, supporting efficient data center hotand cold-aisle designs. Serviceability is enhanced with all customerreplaceable units accessible from the front panel.
●Energy efficiency: The 10-Gbps switches help data centers operate within their space, power, and cooling parameters while reducing their carbon footprints. The switch power supplies are also capable of maintaining 90 percent efficiency at load conditions of as low as 25 percent utilization. This capability allows the switch to make efficient use of power while still being appropriately sized to support the conditions of a full system load.
●Low latency: Cut-through switching enables these switches to support approximately 1 microsecond of port-to-port latency for any packet size with features enabled.
●Intelligent Cisco Switched Port Analyzer (SPAN) and Encapsulated SPAN (ERSPAN): SPAN and ERSPAN can be used for troubleshooting and robust monitoring of traffic. The SPAN and ERSPAN capabilities are nondisruptive, with only extra bandwidth capacity used for SPAN and ERSPAN traffic. Enhancements include more efficient allocation of bandwidth to SPAN and ERSPAN traffic so that any fabric bandwidth not used for data traffic can be allocated to SPAN or ERSPAN traffic. The switch can support up to 31 line-rate SPAN and ERSPAN sessions.
●Flexible buffer management: The 10-Gbps switches support a 25-MB packet buffer shared by every 3 ports of 40 Gigabit Ethernet or every 12 ports of 10 Gigabit Ethernet. The flexible buffer management capability allows dynamic tuning of the sizes of the shared and dedicated buffers in the event of congestion.
●Multicast enhancements: These switches support line-rate Layer 2 and 3 multicast throughput for all frame sizes. They offer optimized multicast replication through the fabric and at the egress point. Support is provided for 32,000 multicast routes and for Internet Group Management Protocol (IGMP) snooping tables in hardware. Multicast enhancements include flow-based hashing for multicast traffic over a port channel and enhanced Bidirectional Protocol-Independent Multicast (Bidir-PIM) support. The switch also supports IP-based forwarding for IGMP snooping.
●Inter-Switch Link (ISL): The Cisco Nexus 5672UP-16G switch supports 16-Gbps Fibre Channel ISLs in a Fibre Channel–only environment or, in the case of FCoE, 40-Gbps ISLs. With six such 40-Gbps links available, the ISLs support bandwidth of 240 Gbps. Improved buffer-to-buffer credits (up to 128) on the new switch now provide support for 16-Gbps Fibre Channel ISLs across distances of up to 16 kilometers.
For a complete list of the latest software features supported on the Cisco Nexus 5600 10-Gbps platform, see the product bulletin at http://www.cisco.com/c/en/us/products/collateral/switches/nexus-5000-series-switches/bulletin- c25-735319.html.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 5 of 24 |
Applications
The Cisco Nexus 5600 10-Gbps platform supports a number of application scenarios, making it a versatile data center option.
Cisco Fabric Extender Architecture: High-Density Fabric Extender Aggregator
Cisco Fabric Extender Technology (FEX Technology) enables you to build a single, modular fabric that extends from Cisco Nexus switches to Cisco Unified Computing System™ (Cisco UCS®) servers, to adapters (Cisco Adapter FEX), and to virtual machines (Cisco Data Center Virtual Machine FEX [VM-FEX]). FEX Technology is based on the emerging standard IEEE 802.1BR. Designing the network using FEX Technology provides flexibility, reduced cabling infrastructure, and a single point of management, helping customers scale their networks. When the 10-Gbps switches are part of a fabric that includes Cisco Nexus 2200 and 2300 platform fabric extenders, you can use these fabric extenders in singleor dual-connected mode, using enhanced virtual port-channel (vPC+) technology to two upstream 10-Gbps switches. Servers and end hosts can connect to single or dual Cisco Nexus 2200 and 2300 platform fabric extenders using network interface card (NIC) teaming when the parent Cisco Nexus 5600 platform 10-Gbps switch has vPC+ enabled.
Following are some common deployment options using the Cisco Nexus 2000 Series (including the 2200 and 2300 platforms) and 5600 10-Gbps platform:
●Rack servers with 100 Megabit Ethernet, 1 Gigabit Ethernet, or 10 Gigabit Ethernet NICs; the fabric extender can be physically located at the top of the rack, and the 10-Gbps switch can reside in the middle or at the end of the row, or the fabric extender and the 10-Gbps switch can both reside in the middle or at the end of the row
●Rack servers with 10 Megabit Ethernet NICs in full duplex mode connected using the Cisco Nexus 2248TP- E Fabric Extender in conjunction with the Cisco Nexus 5600 platform
●Mixed 1 and 10 Gigabit Ethernet environments in which rack servers are running at either speed in the same rack or in adjacent racks
●10 Gigabit Ethernet and FCoE deployments using servers with converged network adapters (CNAs) for unified fabric environments
●10GBASE-T server connectivity with ease of migration from 1 to 10GBASE-T and effective reuse of structured cabling
●1 and 10 Gigabit Ethernet blade servers with pass-through blades
●Low-latency, high-performance computing environments
●Virtualized access
In addition to these options, the 10-Gbps switches provide unique value as a high-density fabric extender aggregation platform. For example, the switches can be used in conjunction with the Cisco Nexus 2348UPQ, 2348TQ, 2332TQ, 2248PQ, 2232PP, 2248TP-E, 2232TM-E, 2232TM, 2248TP, and 2224TP Fabric Extenders as a high-density switching system, consolidating 10 Gigabit Ethernet connections in a single management plane. In addition, a variety of blade fabric extender options can be aggregated into the Cisco Nexus 5600 10-Gbps platform switches using 10 Gigabit Ethernet, providing a single point of management for blade server deployments.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 6 of 24 |
Table 1 lists the fabric extenders that are supported by the Cisco Nexus 5600 10-Gbps platform switches. For more information about the products and the minimum software releases supported, see the Cisco Nexus 2200 and 2300 platform data sheets and release notes.
Table 1. |
Supported Fabric Extenders |
|
|
|
|
Fabric Extender |
Description |
|
|
|
|
Cisco Nexus 2332TQ |
32 x 1/10GBASE-T port host interfaces (SFP+) and up to 4 QSFP+ 10/40 Gigabit Ethernet fabric interfaces; FCoE |
|
|
|
support up to 30m with Category 6a or 7 cables |
|
|
|
Cisco Nexus 2348TQ |
48 x 1/10GBASE-T port host interfaces (SFP+) and up to 6 QSFP+ 10/40 Gigabit Ethernet fabric interfaces; FCoE |
|
|
|
support up to 30m with Category 6a or 7 cables |
|
|
|
Cisco Nexus 2348UPQ |
48 x 1 and 10 Gigabit Ethernet and unified port host interfaces (SFP+) and up to 6 QSFP+ 10/40 Gigabit Ethernet |
|
|
|
fabric interfaces |
|
|
|
Cisco Nexus 2224TP |
24 x 100/1000BASE-T host interfaces and 2 x 10 Gigabit Ethernet fabric interfaces (SFP+) |
|
|
|
|
Cisco Nexus 2248TP |
48 x 100/1000BASE-T host interfaces and 4 x 10 Gigabit Ethernet fabric interfaces (SFP+) |
|
|
|
|
Cisco Nexus 2248TP-E |
48 x 100/1000BASE-T host interfaces and 4 x 10 Gigabit Ethernet fabric interfaces (SFP+; 32-MB shared buffer) |
|
|
|
|
Cisco Nexus 2232PP |
32 x 1/10 Gigabit Ethernet and FCoE host interfaces (SFP+) and 8 x 10 Gigabit Ethernet and FCoE fabric interfaces |
|
|
|
(SFP+) |
|
|
|
Cisco Nexus 2248PQ |
48 x 1/10 Gigabit Ethernet SFP+ host interface and 4 x 40 Gigabit Ethernet (16 x 10 Gigabit Ethernet SFP+) network |
|
|
|
interfaces |
|
|
|
Cisco Nexus 2232TM |
32 x 1/10GBASE-T host interfaces and 8 x 10 Gigabit Ethernet (SFP+) uplink modules |
|
|
|
|
Cisco Nexus 2232TM-E |
32 x 1/10GBASE-T host interfaces and 8 x 10 Gigabit Ethernet (SFP+) uplink modules (lower power consumption and |
|
|
|
improved bit error rate [BER]) |
|
|
|
Cisco Nexus B22HP |
16 x 1/10GBASE-KR internal host interfaces and 8 x 10 Gigabit Ethernet fabric interfaces (SFP+; network interfaces) |
|
|
|
|
Cisco Nexus B22F |
16 x 10GBASE-KR internal host interfaces and 8 x 10 Gigabit Ethernet fabric interfaces (SFP+; network interfaces) |
|
|
|
|
Cisco Nexus B22DELL |
16 x 10GBASE-KR internal host interfaces and 8 x 10 Gigabit Ethernet fabric interfaces (SFP+; network interfaces) |
|
|
|
|
Cisco Nexus B22IBM |
14 x 1/10GBASE-KR internal host interfaces and 8 x 10 Gigabit Ethernet fabric (SFP+; network interfaces) |
|
|
|
|
Large-Scale Fabric (Layers 2 and 3): Leaf and Spine Architecture
Data center designs are evolving, with customers seeking to build large-scale nonblocking fabrics to accommodate different applications, creating patterns of heavy east-west and north-south traffic. The Cisco Nexus 5600 platform 10-Gbps switches are well suited for leaf and spine nodes in a Layer 2 or 3 fabric design. Leaf-and-spine designs using high-density and low-latency switches lead to flatter network architecture, allowing connections that scale from hundreds to more than 10,000 servers with high bidirectional bandwidth and helping ensure low-latency fabric with a low hop count. The spine switches create a nonblocking, low-latency fabric, forwarding packets between leaf switches. The leaf switches provide connectivity to servers. Use of a highly meshed architecture helps ensure the highest possible network availability with little impact on customer traffic in the event of a failure. The 10-Gbps switches can be deployed as Layer 2 or 3 spine or leaf switches, providing a high degree of design flexibility.
Multihop FCoE
Cisco Unified Fabric combines data center and storage networks to deliver a single high-performance, highly available, and scalable network. With the Cisco Nexus 5600 10-Gbps platform switches, Cisco can support end-to- end data center convergence, from the server to storage, by delivering multihop FCoE capability in the data center. The FCoE capability complements the existing FCoE function on the Cisco Nexus 5600 10-Gbps platform. With this broad selection of standards-based FCoE switches, Cisco provides unified fabric support to both the access and core network layers, supporting all storage traffic (FCoE, Small Computer System Interface over IP [iSCSI], and network-attached storage [NAS]) over a simplified infrastructure based on lossless 10 and 40 Gigabit Ethernet.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 7 of 24 |
High-Performance Computing
The Cisco Nexus 5600 10-Gbps platform switches can be deployed as high-density small form-factor (SFF) access-layer switches to consolidate a large number of 10 Gigabit Ethernet servers in deployments that call for only a small number of hops from the server to the upstream network to reduce latency. They have a high density of 10 Gigabit Ethernet ports per rack unit, approximately 1 microsecond of latency port to port for any packet size, integrated line-rate Layer 2 and 3 features, scalability, and integrated data analytics with programmability. They address the needs of high-performance computing (HPC) and high-frequency trading (HFT) environments, for which InfiniBand solutions lack management visibility and high performance of bulk data transfers across traditional applications.
The capability to function in all these capacities helps protect investments in the data center with a deployment model in which additional features can be enabled as they are needed.
Cisco NX-OS Software Overview
NX-OS is a purpose-built data center operating system designed for performance, resiliency, scalability, manageability, and programmability. NX-OS meets Ethernet and storage networking requirements, providing a robust and comprehensive feature set that can meet the demanding requirements of virtualization and automation in present and future data centers. The enhanced Cisco fabric solution allows the transparent integration of the virtual and physical devices on a unified network. In addition, users can use the comprehensive NX-OS service set to create unique innovations for customized solutions. With its MIBs, native XML interface, and command-line interface (CLI) like that of Cisco IOS® Software, NX-OS provides drastically simplified management for the devices in which it runs.
For a complete list of all the features and benefits of NX-OS, see
http://www.cisco.com/en/US/prod/collateral/iosswrel/ps9494/ps9372/data_sheet_c78-652063.html.
Cisco Prime Data Center Network Manager
Cisco Prime™ Data Center Network Manager (DCNM) provides LAN and SAN management capabilities for the Cisco Nexus and Cisco MDS 9000 Families. DCNM provides a GUI that reduces operating expenses (OpEx) compared to traditional CLI methods and allows efficient operation control, monitoring, provisioning, and troubleshooting for your NX-OS devices. The main features include the following:
●Unified fabric visibility and topology display with VMware vSphere integration shows the connectivity from the virtual machine to the VMware ESX host and to the switch and the storage array.
●Event aggregation and filtering helps you quickly find the information you need and identify network problems.
●Deployment wizards and user-modifiable templates help you implement best practices.
●Role-based access control (RBAC) helps secure devices and provide appropriate delegation.
●Integrated domain dashboards, health monitoring, reporting, change tracking, and user auditing provides comprehensive management capabilities.
●Trend monitoring of ports and traffic allow you to optimize your existing resources and anticipate new resource requirements.
© 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. |
Page 8 of 24 |
Loading...