Full English Description for Cisco N9K-C9336C-FX2
1. Part Number Naming Breakdown
-
N9K: Cisco Nexus 9000 cloud-scale fixed data center switch product family
-
C9336: All-QSFP28 high-density spine/leaf chassis of Nexus 9300 FX2 second-generation Cloud Scale ASIC platform
-
C: Universal QSFP28 multi-rate ports supporting 40G / 100G native speeds with full breakout capability
-
FX2: Second-gen upgraded Cloud Scale ASIC, larger shared packet buffer, expanded LPM route table capacity, optimized RoCEv2 lossless transport, wire-rate MACsec on all ports, high-performance unsampled streaming telemetry, dual operational modes: standalone NX-OS and ACI intent-based fabric mode
-
Form factor: Compact 1RU standard rack-mount chassis
2. Short Sales One-Liner (For BOM, Quotation & Datasheet Header)
Cisco Nexus N9K-C9336C-FX2, 1RU rack-mount all-optical dual-purpose spine/leaf data center switch, 36 universal multi-rate QSFP28 10/25/40/100G ports, 7.2 Tbps non-blocking switching capacity, dual hot-swappable Platinum-efficiency redundant power supplies and three N+1 variable-speed fan trays, dual-mode standalone NX-OS or ACI fabric operation, optimized for data center core spine interconnection, GPU/HPC low-latency clusters, high-density leaf aggregation, cross-pod fabric border nodes and hyperscale multi-tenant cloud deploymentsCisco.
3. Complete Product Overview Paragraph
The Cisco Nexus N9K-C9336C-FX2 is a compact 1RU fixed-port Layer 2/3 all-optical switch built on the upgraded Nexus 9300 FX2 Cloud Scale ASIC platform, uniquely designed to operate flexibly as either a high-capacity spine switch or high-speed leaf switch for enterprise, hyperscale cloud, AI GPU and high-performance computing (HPC) fabrics, including Cisco ACI Application Centric Infrastructure environments. It features a full 36-port QSFP28 layout without copper downlinks, delivering complete wire-rate non-blocking forwarding with total switching bandwidth of 7.2 Tbps and a forwarding throughput of 2.8 Bpps for minimum 64-byte Ethernet framesCisco.
Every one of the 36 QSFP28 ports supports independent flexible breakout configurations: native 100G, native 40G, 2×50G NRZ, 4×25G or 4×10G via passive breakout cables and QSA adapters, enabling seamless fabric speed migration across 10G, 25G, 40G and 100G generationsCisco. Limited 1G support is available via QSA adapters on ports 7–30 only; ports 1–6 and 33–36 do not support 1G transceivers. All ports integrate hardware wire-rate 128/256-bit MACsec encryption natively, eliminating separate security appliance overhead for fabric inter-switch linksCisco.
Powered by FX2-series Cloud Scale silicon, this platform delivers an expanded LPM route table capacity up to 896,000 IPv4/IPv6 entries, a 40MB shared dynamic packet buffer to absorb burst traffic, native hardware acceleration for EVPN/VXLAN overlay routing, and optimized RoCEv2 lossless congestion control for low-latency GPU and storage workloads. High-performance unsampled streaming telemetry is fully integrated with Cisco Tetration Analytics for real-time end-to-end fabric observabilityCisco.
The chassis integrates dual independent management interfaces: dedicated out-of-band RJ45 Ethernet and SFP+ fiber management uplink ports, plus mini-USB RS232 console and single USB Type-A port to support POAP zero-touch provisioning, offline configuration backup and firmware upgrades. Dual hot-swappable power supplies and three N+1 redundant fan trays eliminate single points of hardware failure, with bidirectional airflow options (PI port-side intake / PE port-side exhaust) selectable at order time to match diverse rack thermal layout standards. This switch supports two independent operating modes: modular standalone Cisco NX-OS for traditional routed data center architectures, or native ACI fabric mode for policy-driven automated multi-tenant cloud infrastructure. Its all-100G optical design maximizes bandwidth density for core spine layers and high-speed GPU leaf racks within a minimal 1RU rack footprint.
4. Detailed Hardware Interface Specifications
Universal QSFP28 Ports (36 × Full Multi-Rate Slots, all ports MACsec-capable)
-
Supported native speeds: 40GBASE, 100GBASE
-
Full breakout options per port:
1 × 100G, 1 × 40G, 2 × 50G NRZ, 4 × 25G, 4 × 10G via breakout cables / QSA adapters
-
1G limitation: 1G transceivers only supported via QSA on ports 7–30; ports 1–6 & 33–36 block 1G operation
-
Compatible transceivers & cables: QSFP28 100G, QSFP+ 40G, SFP28 25G, SFP+ 10G, passive short-reach DAC twinax cables
Management & Auxiliary Ports
-
1 × 10/100/1000BASE-T RJ45 out-of-band management port
-
1 × SFP+ multi-rate fiber out-of-band management uplink port
-
1 × Mini-USB RS232 console port for local CLI access
-
1 × USB 2.0 Type-A port for firmware upload, configuration backup and POAP deployment
Power & Cooling
-
Dual hot-swappable power supply options: 750W AC, 1100W AC, 1100W DC, 1100W HVAC/HVDC (1+1 load-sharing redundancy, live hot-swap without traffic interruption)
-
Three variable-speed fan trays (2+1 N+1 redundancy), bidirectional airflow PI / PE configurable at ordering (burgundy = PI intake, blue = PE exhaust)
-
Power efficiency: 80Plus Platinum certified
-
Typical AC power draw: 336.6W; maximum AC power draw: 718.7W
-
Operating temperature range: 0°C ~ 40°C under full load; non-operating storage range: -40°C ~ 70°C
-
Humidity tolerance: 5%–90% non-condensing operating; max operating altitude: 4000 meters
Physical Form Factor
-
Dimensions: 1RU standard 19-inch rack chassis, 4.4cm (H) × 43.9cm (W) × 62.2cm (D)
-
Empty weight (without PSUs and fans): 8.5 kg / 18.8 lb
-
Compliance: RoHS-6, NEBS Level 3, global safety and EMC certifications
Core Hardware Performance Benchmarks
-
Total switching capacity: 7.2 Tbps full duplex non-blocking
-
Forwarding rate: 2.8 Bpps (64-byte frame wire-rate)
-
Shared dynamic packet buffer: 40MB for cross-port burst traffic absorption
-
MAC address table size: 256,000 MAC forwarding entries
-
IPv4 / IPv6 unicast LPM route table capacity: 896,000 routing entries
-
VLAN support: Up to 4094 standard IEEE 802.1Q tagged VLANs
-
System memory: 24GB DRAM + 128GB onboard SSD storage
-
Typical cut-through switching latency: Less than 1 microsecond for minimum-size Ethernet frames
5. Core Software Feature Suite (Dual NX-OS / ACI Operating Modes)
Layer 2 & Layer 3 Routing
Static routing, RIP, OSPFv2/v3, EIGRP, BGP, VRRP, PIM sparse/dense multicast; FCoE not supported on this all-high-speed optical chassis
Cloud Fabric Overlay
Hardware-accelerated EVPN/VXLAN symmetric & asymmetric routing, multi-tenant VXLAN segmentation, fabric multicast optimization, RoCEv2 lossless congestion control for GPU/HPC low-latency workloads
Enterprise & Fabric Security
802.1X port authentication, MAB MAC bypass authentication, DHCP Snooping, Dynamic ARP Inspection, CoPP control-plane policing, hardware 128/256-bit MACsec encryption on all 36 data ports, TrustSec SGT zero-trust micro-segmentation
Monitoring & Telemetry
High-performance sampled/unsampled IPFIX streaming telemetry, ERSPAN remote port mirroring, local SPAN/VSPAN, real-time per-port/per-queue buffer utilization visibility, full Cisco Tetration flow export compatibility
Automation & DevOps
POAP zero-touch provisioning, NX-API REST JSON/XML APIs, embedded Linux Bash shell, native Python scripting, Ansible/Puppet/Chef integration, containerized application hosting, hot patching and ISSU in-service software upgrade without traffic downtime
High Availability
Modular fault-isolated process architecture, stateful process restart, redundant control plane, hot-swappable power supply and fan FRUs
6. Typical Deployment Use Cases
-
Core spine switch for large-scale leaf-spine data center fabrics (40G/100G inter-leaf interconnection)
-
Low-latency leaf node for GPU virtualization, AI training and HPC parallel computing clusters
-
Cross-pod border spine switch connecting multiple independent data center pods
-
High-speed leaf aggregation node for all-optical 25G/100G server racks without copper cabling
-
Fabric core node for medium and large-scale ACI Application Centric Infrastructure multi-pod environments
-
Hyperscale IaaS cloud core switch supporting multi-tenant isolated fabric segmentation with wire-rate MACsec encryption
7. Datasheet Short Summary Paragraph
The Cisco Nexus N9K-C9336C-FX2 is a 1RU FX2-generation all-optical dual-purpose spine/leaf cloud data center switch equipped with 36 multi-rate 10/25/40/100G QSFP28 universal ports, delivering 7.2 Tbps non-blocking switching performance with dual Platinum redundant power and three redundant fan trays. Supporting both standalone modular NX-OS and ACI intent-based fabric operational modes, it features upgraded FX2 ASIC with expanded L2/L3 routing scale, hardware EVPN/VXLAN overlay, full-port wire-rate MACsec encryption and optimized RoCEv2 lossless transport, serving as a high-bandwidth compact core spine or high-speed leaf aggregation switch for hyperscale cloud, GPU/HPC and large virtualized enterprise data centers built around 40G/100G optical fabric infrastructureCisco.
|
